international journal of computing and ict research issue 2 2009-2.pdf · international journal of...

International Journal of Computing and ICT Research, Vol. 3, No. 2, December 2009.

Contents Volume 3, Issue 2, December 2009.

ISSN 1818-1139 (PRINT), ISSN 1996-1065 (ONLINE) Technology and Academic Dishonesty – Part II: A Focus on Academicians and Other Researchers Joseph M. Kizza – Editor-in-Chief E-governance and Online Public Service: The Case of a Cyber Island Taruna Shalini Ramessur Correlates of Computer Attitude among Secondary School Students in Lagos State, Nigeria Adebowale, O.F, Adediwura, A.A., Bada, T. A. A Mobile Real Time Interactive Communication Assistant for Cerebral Palsy Lau Bee Theng, Low Tiong Kie and Hii Kiing Shi Utilizing Semantic Web as Communication Protocol in Faded Information Field (FIF) Architecture for Information Retrieval and Dissemination M. Ahsan Chishti, Shaima Qureshi, A. H. Mir, Shariq Haseeb and Iftikhar Ahmad An Information Databank Framework for the health Care Industry in Nigeria Olutola M. Obembe and Oluruntoba S. Ogundele An Effective Algorithm Using Similarity Based Technique for Browsing Image and Databases Rekha B Venkatapur, V.D.Mytri and A.Damodaram

International Journal of Computing and

ICT Research

2


International Journal of Computing and ICT Research Editorial Board Editor-in-Chief: Prof. Joseph M. Kizza, Department of Computer Science and Engineering College of Engineering and Computer Science The University of Tennessee-Chattanooga, 615 McCallie Avenue, Chattanooga, Tennessee, USA [email protected] Managing Editors:

Computer Science Prof. Anthony Rodrigues, University of Nairobi, Kenya

Information Technology Prof. Shushma Patel, London South Bank University, UK

Information Systems Prof. Ravi Nath, Creighton University, Nebraska, USA

Computer Engineering Prof. H.N Muyingi, Forthare University , South Africa

Software Engineering Prof. P.K. Mahanti, University of New Brunswick, Canada

Data Communication and Computer Networks Prof. Vir Phoha, Louisiana Tech, USA

ICT for Sustainable Development Prof. Kathy Lynch, University of the Sunshine Coast, Queensland, Australia Production Editor: Book Review Editor: Prof. Timothy Waema, School of Computing and Informatics, The University of Nairobi, Kenya Journal Editorial Office: The International Journal of Computing and ICT Research Makerere University P.O. Box 7062, Kampala, Uganda. Tel: +256 414 540628 Fax: +256 414 540620 Email: [email protected] Web: http://www.ijcir.org

3


Volume 3, Issue 2. December 2009. The International Journal of Computing and ICT Research Makerere University P.O. Box 7062, Kampala, Uganda. Tel: +256 414 540628 Fax: +256 414 540628 Email: [email protected] Web: http://www.ijcir.org

International Journal of Computing and ICT Research

4


Volume 3, Issue 2, December 2009.

Table of Contents

Technology and Academic Dishonesty – Part II: A Focus on Academicians and Other Researchers Joseph M. Kizza – Editor-in-Chief …………………………………………………………………………7

E-governance and Online Public Service: The Case of a Cyber Island

Taruna Shalini Ramessur………………………….………………………………………………12 Correlates of Computer Attitude among Secondary School Students in Lagos State, Nigeria

Adebowale, O.F, Adediwura, A.A., Bada, T. A. ………………………………………………………..20 A Mobile Real Time Interactive Communication Assistant for Cerebral Palsy

Lau BeeTtheng, Low Tiong Kie and Hii Kiing Shi …………………………..……….…………………31

Utilizing Semantic Web as Communication Protocol in Faded Information Field (FIF) Architecture for Information Retrieval and Dissemination

M. Ahsan Chishti, Shaima Qureshi, A. H. Mir, Shariq Haseeb and Iftikhar Ahmad …………41 An Information Databank Framework for the health Care Industry in Nigeria

OlutolaM. Obembe and Oluruntoba S. Ogundele……………………….………………….………….47 An Effective Algorithm Using Similarity Based Technique for Browsing Image and Databses

Rekha B Venkatapur, V.D.Mytri and A. Damodaram………….…………………58

5


Book Reviews

Every issue of the journal will carry one or more book reviews. This is a call for reviewers of books. The book reviewed must be of interest to the readers of the journal. That is to say, the book must be within the areas the journal covers. The reviews must be no more than 500 words. Send your review electronically to the book review editor at: [email protected]

6


International Journal of Computing and ICT Research The IJCIR is an independent biannual publication of Makerere University. In addition to publishing original work from international scholars from across the globe, the Journal strives to publish African original work of the highest quality that embraces basic information communication technology (ICT) that will not only offer significant contributions to scientific research and development but also take into account local development contexts. The Journal publishes papers in computer science, computer engineering, software engineering, information systems, data communications and computer networks, ICT for sustainable development, and other related areas. Two issues are published per year: June and December. For more detailed topics please see: http://www.ijcir.org.

Submitted work should be original and unpublished current research in computing and ICT based on either theoretical or methodological aspects, as well as various applications in real world problems from science, technology, business or commerce.

Short and quality articles (not exceeding 20 single spaced type pages) including references are preferable. The selection of journal papers which involves a rigorous review process secures the most scholarly, critical, analytical, original, and informative papers. Papers are typically published in less than half a year from the time a final corrected version of the manuscript is received.

Authors should submit their manuscripts in Word or PDF to [email protected]. Manuscripts submitted will be admitted subject to adherence to the publication requirements in formatting and style. For more details on manuscript formatting and style please visit the journal website at: http://www.ijcir.org.

7


Technology and Academic Dishonesty – Part II: A Focus on Academicians and Other Researchers

PROF. JOSEPH M. KIZZA∗ Editor-in-Chief Abstract

The rapid technological developments of the last twenty years have been a catalyst in creating new avenues of academic dishonesty for both academician and other researchers. Although this vice has ravaged the academic and research communities for years, it has reached epidemic proportions because of the Internet. Academicians and researchers at every level are submitting works verbatim, most of such works downloaded from the Internet and falsifying and sexing up data. This article is part II in our two part series of articles discussing academic dishonesty. In this part we continue our discussion academic dishonesty but this time focusing on academicians and other researchers highlighting how the Internet has impacted these practices, the efforts being made to curb them and the effects they are having on society.

IJCIR Reference Format: Kizza, Joseph. M. Technology and Academic Dishonesty – Part II: A Focus on Academicians and Other Researchers. International Journal of Computing and ICT Research, Vol. 3, No. 2, pp. 7-11. http://www.ijcir.org/volume3-number2/article1.pdf. ______________________________________________________________________________________ 1. INTRODUCTION

In part I of this series which appeared in the in the Special Issue of International Journal of Computing and ICT Research Number 1, 2009, we discuss academic dishonesty among students, focusing on what it is, how wide spread it is and ways to counter it. In this part II, a continuation of our discussion of academic dishonesty, let us focus on both faculty and other researchers. These popular vices and a less known but more pervasive one, self-plagiarism, are discussed here highlighting how the Internet has impacted these practices, efforts being made to curb them, and the effects they are having on society.

The rapid technological developments of the last twenty years have been a catalyst in creating new aids in research and avenues for faculty to publish their works. However, technological developments have also given impetuous and a new fire into an old vice of academic dishonesty. Although this vice has ravaged the academic community for years, it has reached epidemic proportions because of the Internet. Academicians at every level are submitting works verbatim, most of such works downloaded from the Internet. Faculty and scientists, under the weight of publish or perish, or seeking a minute of fame, can very easily sex up data or publish works of others as their own.

Lifting of others works to benefit and expound your minute intellectual might and/or to blemish and sex up otherwise less appealing research is to damage the academic research integrity and commitment which is detrimental to the trustworthiness of the whole research process by the greater scientific community. Do not forget that trustworthiness in our research is the engine that drives the research

∗ Author’s Address: Joseph M. Kizza, Department of Computer Science and Engineering, The University of Tennessee-Chattanooga, Chattanooga, TN 37403, USA. [email protected]. "Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than IJCIR must be honored. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee." © International Journal of Computing and ICT Research 2009. International Journal of Computing and ICT Research, ISSN 1818-1139 (Print), ISSN 1996-1065 (Online), Vol.3, No.2 pp. 7-11, December 2009.

8


community to have the vigor to innovate and flourish. Researcher trustworthiness is the glue that binds us together to ensure a climate that promotes confidence and trust in our research findings, encourages free and open exchange of research materials and new ideas and brings about respect of the intellectual contributions of all researchers wherever they are.

The U.S. based National Science Foundation (NSF) defines research misconduct as "fabrication, falsification, plagiarism, or other practices that seriously deviate from those that are commonly accepted within the scientific community for proposing, conducting or reporting research. These are by no means new vices, but they have been given momentum, visibility, made easy, seemingly decriminalized, and at times even legitimized through the use of the Internet.

2. TYPES OF ACADEMIC DISHONESTY

There are several types of academic dishonesty including: • Plagiarism - the act of using another person’s ideas or writings as one’s own without

acknowledgement for their use • Fabrication - making up results and recording or reporting them • Falsification of information, data or citation through authorized or unauthorized access and

presenting that data as original; presenting information or data that were not gathered in accordance with standard guidelines defining the appropriate methods for collecting or generating such data and failing to include an accurate account of the method by which the data were gathered or collected

2.1 Plagiarism The word plagiarism comes from a Latin verb meaning to kidnap. It describes the process of intentionally or even unintentionally “kidnapping” which in every aspect is stealing others people’s work and intellectual property without their consent. On academic institutions and research sites across the globe, this vice has been sky-rocketing ever since the start of the Internet.

Although the known statistics of academicians and researchers plagiarizing are low, there is a growing problem within the academic ranks propelled by pressure for publication as prerequisite for promotion and tenure. The problem is not only limited to groups of universities based on their academic rankings or regions of the country or world but it cuts across the board in university rankings, geographical regions, faculty and researchers ages and rankings. For example, there is a small but growing number of well established scholars who have not only resorted to sexing up ideas but are doing cut and paste. According to a report in the Chronicle of Higher Education, of Friday, October 25, 2002, Michael Bellesiles, a well renowned history professor at Emory University resigned from his position. His resignation was sparked by allegations of plagiarism in his award winning book Arming America. Emory launched an investigation into his misconduct. The findings documented in the report state that Bellesiles's "carelessness in the gathering and presentation of archival records" raised questions about his "scholarly integrity" [CHRONICLE]. Following these allegations and findings of Emory University, Columbia University rescinded the 2001 Bancroft Prize in American History and Diplomacy for his book, Arming America: The Origins of a National Gun Culture. The trustees determined that Bellesiles' book "'had not and does not meet' the standards associated with the prestigious prize" [CHRONICLE]. In the same year (2002), renown American historians Stephen Ambrose and Doris Kearns Goodwin both best- selling historians were accused of plagiarism and improper citations for passages of their books apparently taken from other writers. 2.2 Fabrication and Falsification Fabrication is the practice of “doctoring” data or results to suite a “wished” , expected or prescribed outcome or reporting. Falsification on the other hand is the process of “sexing up” existing data to fit a specific study. Both of these schemes are dangerous to research because they alter the reliability of data that may be used in systems that are sometimes vital to life putting thousands of lives in danger and sometimes death. Ideally, the research community would not expect fabricated and falsified data. However, we do not live in Utopia. We live in a real world of money making schemes and competition. Studies have shown that fabrication and falsification are widespread practices through all sectors of society including academia. Just as plagiarism diminishes the value of the works and tarnishes the integrity and ethics of the author, fabrication and falsification of research data diminishes the status of the research and

9


may create a dangerous situation for those who use such data. For example pharmaceutical and medical research fraud has potentially devastating implications

because doctors base treatment on published research. It was reported in the U.S. sometime back that technology has enabled some unscrupulous individuals to manufacture imitation drugs and labels (imitations of expensive drugs like those for cancer treatment) with disastrous consequences.

In a story by David Burge of CNSNews.com Academic Affairs and Satire of March 15, 2002, in September 2000, a renowned Princeton psychology professor Victor Agee published a monograph in the Journal of Social Psychology measuring test-cheating behavior among undergraduate students at Columbia. The results of Agee's cheating study were widely cited in the popular press, and he quickly gained fame as a television talk show panelist.

In June 2001, Agee was suspended after two Cornell University statisticians, Dipak Sharma and Peter Millard, published a paper in the journal Biometrika accusing Agee of data manipulation and outright falsification. The results of their study were widely cited in the popular press, and the pair quickly gained fame as television talk show panelists.

According to Yale historian Alan Levin, however, Sharma and Millard's expose of Agee was in large part also plagiarized from a 1993 University of Michigan paper chronicling fraud in a Duke study of cheating at the University of California, which itself appears to have been plagiarized by Agee. Levin was later forced to cancel several television talk show appearances to defend himself against charges that he was engaged in an inappropriate relationship with NYU professor Olivia Nolan, editor of The Journal of University Ethics.

This account throws light on the dimension of the data fabrication and falsification in academia. However, this is not limited to academia alone. It is widespread in the greater research community. 2.3 Self-Plagiarism Self-plagiarism occurs when an author re-uses portions of his or her previous published work in subsequent works. Sometimes a few bits and pieces are stitched and pasted together scattered all throughout the paper; sometimes a whole paper is copied with a simple change of the title. According to Christian Colleberg and Stephen Kobourov, self-plagiarism is more prevalent than other forms of academic dishonesty [COLLEBERG, CHRISTIAN and STEPHAN KOBOUROV], and one of the reasons might be because it is difficult to determine what is fair use and what constitutes self-plagiarism.

There are confusing and sometimes uncaring policies on self-plagiarism on the part of major professional organizations and publishing houses. For example IEEE Computer ( http://www.computer.org/portal/site/transactions/menuitem.eda2ca84d8d67764cfe79d108bcd45f3/index.jsp?&pName=transactions_level1&path=transactions/tmc/mc&file=author.xml&xsl=article.xsl&) states: “Papers previously published in conference proceedings, digests, preprints, or records are eligible for consideration provided that the author informs the Transactions Assistant at the time of submission and that the papers have undergone substantial revision. The question regarding concurrent submission appears on Screen 1 in Manuscript Central.” ACM (http://www.acm.org/pubs/sim_submissions.html) also has the following statement: “ the paper has been substantially revised (this generally means that at least 25% of the paper is material not previously published; however, this is a somewhat subjective requirement that is left up to each publication to interpret”. Too many, substantial revision means anywhere around at least 25% new material in the paper. But again this is subject and it is usually left to the author or the publishing agent to decide. There is also little agreement among professionals and academicians what this means. Colleberg et al list the different types of self -plagiarism as follows (the word re-use is used here because it is soft) [COLLEBERG, CHRISTIAN and STEPHAN KOBOUROV]:

• Textual re-use - Incorporating text/images/other forms of previously published works in a new work

• Semantic re-use – Incorporating ideas from published works in a new work • Blatant re-use - Incorporating text/ideas of previously published works in a new work almost

verbatim • Selective re-use – Borrowing bits and pieces from previously published works in a new work • Incidental re-use - Incorporating text/ ideas not directly related to the new ideas in the current

document • Re-use by cryptomnesia - Incorporating text/ideas from previously published works in a new

work unaware of its existence

10


• Opaque re-use - Incorporating text/ideas from previously published works in a new work without acknowledging its existence

• Advocacy re-use - Incorporating text/ideas from previously published works in a new work when writing for a different audience

While most agree that blatant and opaque re-uses are ethically and legally wrong, there is disagreement of other forms of re-use especially advocacy. Colleberg et al suggest some solutions to self-plagiarism [COLLEBERG, CHRISTIAN and STEPHAN KOBOUROV]:

• Conference organizers and journal publishers to clearly state what practices will be considered self-plagiarism

• Publishing houses to take more of the burden of checking self-plagiarism • Making the price of self-plagiarism high.

3. CAUSES OF ACADEMIC DISHONESTY

For academicians and to some extent other researchers, the pressures caused by the policy of “publish or perish” is the main culprit causing a bigger percentage of academic dishonesty. The tenure system is the bedrock of academic longevity and rewards. However, without publishing in quality journals, no academician is any self-respecting institution of higher learning can secure the cherished tenure. In non-academic research institution, it is promotions, pay rises, and recognition among peers that drives publication and subsequently academic dishonesty.

4. EFFECTS OF ACADEMIC DISHONESTY

Academic dishonesty and sexing up of ideas have many very negative effects on society. Such effects include ethical and moral decay as people’s actions become routine and eventually mainstreamed. Florence King notes in the National Review that “once plagiarism becomes an officially designated addiction, its sufferers will not only be forgiven, but admired.” [KING, FLORENCE] This is the way it has been working and it will continue to work especially with the public so dependent on technology and still mesmerized by it. Another effect is highlighted by Elliott J. Gorn, professor of History at Purdue University. According to Gorn,” beyond the ethical issues lays a serious challenge to the historian's craft, for the foundation of [historical] narratives, the bedrock of our interpretations, are the facts we uncover in primary documents. Put another way, it is not only our skill at interpreting historical sources, but also our integrity in presenting them -- with all of their contradictions and complexity -- that authorizes us to bear witness to the past.” [GORN, ELLIOTT]. Further academic dishonesty erodes our ability as academicians and researchers to speak with authority of our disciplines. There are also serious psychological effects on individuals who commit these vices as they tend to become reclusive and self-isolated which may lead to dangerous consequences on the individual and society. Finally, it causes serious security and safety problems to society because essential products like software and hardware systems which may be made from fabricated and falsified data may become mainstreamed and may cause harm to life.

5. GOING AFTER STUDENTS’ ACADEMIC DISHONESTY:

Fighting these vices among academics and researchers needs to have a multi-faceted approach that includes policies and detection tools. 5.1 Policies A cheating and plagiarism policy would include a definition of the vice, a list of items that would constitute the offense resulting from the vice, the enforcement mechanisms available to university and research institutions administrators and the penalties for the offenders. 5.2 Plagiarism Detection Software & Services Because the numbers of students usually taught by one instructor may be big, it may be a challenge to teachers to find less tedious ways to find cheaters. Fortunately the technology that is spearheading the growth of these vices is the same technology that is providing the remedy for them. With the growth of plagiarism, fabrication, falsification, and cheating, there is a growing base of technologies to fight them.

11


There are several software tools ( sometimes free) like Viper on (http://www.scanmyessay.com/viper-plagiarism-scanner.php), WCopyfind on (http://plagiarism.phys.virginia.edu/Wsoftware.html), pl@giarism (http://www.plagiarism.tk). Other sorces include:

• EVE2: http://www.canexus.com/eve/index.shtml • WordCheck Systems: http://www.wordchecksystems.com/ • iParadigms: http://www.iparadigms.com/ • Glatt Plagiarism Program: http://www.plagiarism.com/

Depending on the nature of the suspected work, a teacher can use the software to compare the paper against textbooks, literature, works of other students, or the products of online paper mills--sites where students can download papers. Although we cannot explore all efforts being made to curb academic dishonesty here because of time limits, literature on the topic abounds. You can find useful information on this topic, guess where - Internet

6. REFERENCES

CHRONICLE OF HIGHER EDUCATION. “ Bellesiles Resigns From Emory After University Report Questions His Research for Book on Guns” http://chronicle.com/daily/2002/12/2002121604n.htm

CHRONICLE OF HIGHER EDUCATION. “Columbia U. Rescinds Bancroft Prize Awarded to Michael Bellesiles for Book on Gun Ownership”http://chronicle.com/daily/2002/12/2002121604n.htm

GORN, ELLIOTT. “Why Are Academics Ducking the Ellis Case?” OAH Newsletter. KING, FLORENCE ” The Misanthrope's Corner - Doris Kearns Goodwin and the latest plagiarism scandal

- Brief Article – Column”. National Review. April 8, 2002. COLLEBERG, CHRISTIAN AND STEPHAN KOBOUROV. “ Self-Plagiarism in Computer Science”.

Communication of the ACM, Vol. 48, No. 4, 2005.

12


E-Governance and Online Public Service: The Case of a Cyber Island

TARUNA SHALINI RAMESSUR* University of Technology, Mauritius Abstract

One of the central research questions emerging from the favorable and critical views on eGovernance is how such a new mode of governance has impacted on service delivery in the public sector. This dimension is crucial, because what matters most is whether the adoption of eGovernance has been able to improve service delivery, one of the core functions of Governments — based on quality, processes and operations. This article explores these issues by analyzing a specific eservice (online application for learner’s licence) provided by the Government of Mauritius, a country which represents one of the leading advocates of eGovernance in Sub Saharan Africa. The results from the survey undertaken indicate that the most important effect of eGovernance on the application for learner’s license is speeding up of processes and better quality of service in terms of responsiveness and reliability but not in terms of access and security. As far as the process is concerned eGovernance has lead to personalized service but has not covered all the physical aspect of the service. In terms of operations the resulting effect was the modernisation of the service. General Terms: Human Factors, Reliability, Security Additional key Words and Phrases: e-Governance, service delivery, Mauritius IJCIR Reference Format: Taruna Shalini Ramessur, E-governance and Online Public Service: The Case of a Cyber Island. International Journal of Computing and ICT Research, Vol. 3, No. 2, pp. 12 - 19. http://www.ijcir.org/ volume3-number2/article2.pdf. 1. INTRODUCTION

The actual trend today is towards simultaneous globalisation and localisation and as such the State is increasingly torn apart between the global and the local, especially in the case of one of its core functions that is service delivery. Service delivery is indeed more and more taking place below and above nation-state levels, mainly because the (private) operators delivering such services are in the process of restructuring at these levels. Moreover due to the fact that service delivery in the public sector is increasingly being outsourced or subcontracted, it requires a complex governance structure. Such trends have contributed to even bigger pressure on the State to improve service delivery to citizens and increased the acceptance that achieving excellence in customer service is just as critical for the public sector as it is for private companies. As a result in many parts of the world Governments are having recourse to eGovernance to achieve this goal. Egovernance is a new term in a family of a rapidly expanding vocabulary of e-prefixed terminologies reflecting the expanding role of ICT in society. It has its origins in the emergence of Internet-based

* Author’s Address: Taruna Shalini Ramessur, University of Technology, La Tour Koenig Pointe-aux-Sables, Mauritius. Tel: +230 234 6535; Fax: +230 234 6219; Email : [email protected] "Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than IJCIR must be honored. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee." © International Journal of Computing and ICT Research 2009. International Journal of Computing and ICT Research, ISSN 1818-1139 (Print), ISSN 1996-1065 (Online), Vol.3, No.2, pp. 12-19, December 2009.

13


applications that enable electronic delivery of information and services in both business and government. There are various definitions of e-governance but for the purpose of this paper the term eGovernance is defined as the process of using information technology for automating both the internal operations of the government and its external interactions with citizens and other businesses. The three main target groups that can be distinguished in e-governance concepts are government, citizens and businesses/interest groups. The external strategic objectives focus on citizens and businesses and interest groups, the internal objectives focus on government itself. 1.1 Purpose With the above discussion in mind, the purpose of this paper becomes: assessing the impact of eGovernance through eservice on service delivery. In order to gain knowledge necessary for accomplishing the sated purpose, the following research questions will be looked into. 1. How eservice has affected quality of service delivery? More specifically the quality effects are assessed in terms of:

• reliability • responsiveness • access • ease of use • attentiveness • credibility • security

2. How eservice has influenced the process of service delivery? As far as the processes are concerned, the effect of eGovernance will be examined in the following ways:

• attending to the relationship between the administration and the citizen at the transaction level.

• drawing the State closer to the market 3. How eservice has impacted on the operation of service delivery? Finally the operation effects will be assessed in terms of:

• bringing Government closer to citizens • modernising Public Services • reducing opportunities for trivial fraud at different point of service delivery • increasing mechanisms to construct more accountability and transparency in the

Public Sector

2. LITERATURE REVIEW

The cyber-optimists believe that eGovernance holds great promise for the delivery of many types of public services from housing and welfare benefits to community health care and the electronic submission of tax returns, reconnecting official bureaucrats with citizen/customers*. The Internet can serve multiple functions: disseminating information about the operation of government as well as public services, facilitating public feedback mechanisms like emails to government agencies, enabling more direct participation into the decision making process including consultation exercises at local level, and providing direct support for the democratic process, such as the efficient administration of electoral registration or online voting†. There is widespread concern that the public has lost faith in the performance of the core

* See Stavros Zouridis and Victor Bekkers. 2000. ‘Electronic Service Delivery and the democratic relationships between government and its citizens.’ In Jens Hoff, Ivan Horrocks and Pieter Tops. Eds. Democratic Governance and New Technology. London: Routledge; Rob Atkinson. 2000. ‘Creating a Digital Federal Government.’ IMP: Information Impacts Magazine. October. www.cisp.org/imp. † See Elisabeth Richards. 1999. ‘Tools of Governance’ and Eileen Milner. 1999. ‘Electronic Government: More than Just a Good Thing?’ In Digital Democracy: Discourse and Decision Making in the Information Age. Ed. Barry N. Hague and Brian D. Loader.

14


institutions of representative government, and it is hoped that more open and transparent government and more efficient service delivery could help restore public confidence*. In developing societies, the Internet can potentially help with the multiple challenges facing the effective delivery and administration of basic government services such as health and education, especially given the global reach that the technology provides, connecting medical professionals, local officials and university teachers in Oslo, Cambridge and Geneva with those in Nepal, Bangalore, and Havana.

Egovernance makes information available on government operations and public services, facilitates public feedback or reaction and allows more direct participation by the ordinary citizen in decision-making [Heeks, 2001b; Norris, 2001]. The eGovernance movement not only promises higher quality and better delivery of services and a greater realization of entitlements, it also claims to offer stronger bonds between public servants and citizens based on transparency and accountability [Heeks, 2001a].

Schware [2000] emphasizes that egovernance provides equal access to government and speedy and transparent responses from public servants. In addition, egovernance provides a wider opportunity for public servants to interact directly with the public in the process of receiving feedback from citizens and responding to their queries and complaints through electronic means. For Ghere and Young [1998], public agencies now have to justify their decisions based on feedback from the people and conduct their business in public. The main rationales behind opting for egovernance are that egovernance will reduce costs and delays in delivering services, expand citizens’ access to public sector information, reinforce innovation in public agencies, increase transparency and public accountability, weaken authoritarian tendencies and strengthen civil society and democracy [Pardo, 2000; Heeks, 2001a; Norris, 2001].

However the cyber pessimists believe that the use of IT in governance may worsen inequality in access to government services due to the lack of an adequate infrastructure, unequal ownership of computers, language constraints, and so on [UNDP, 1999; Singh, 2000; Levine, 2001]. There is also a concern that egovernance may disempower citizens by individualizing them, eroding their common bonds and endangering their privacy [Ghere and Young, 1998; Wachbroit, 2001]. For the critics, instead of a citizen–administration relationship based on equality and accountability, egovernance may strengthen a top-down bureaucratic process by posting information about the structures and functions of public agencies and reinforcing the existing mode of interaction through documents and reports [Norris, 2001].

Moreover it is argued that egovernance may not only increase the power of bureaucratic experts in relation to elected political leaders, it may also lead to the politicization of the overall bureaucracy. If the information-expert bureaucrats become too influential in relation to elected political representatives, it may undermine their accountability to these elected politicians. In other words, under egovernance, the nature of the relationship between politicians and public servants may have changed from one based on neutrality and accountability to one of a fused power structure with the dominance of bureaucrats empowered by information expertise. In this regard, Daly [2000] makes a general observation that the use of the internet in governance has enhanced the dominance of nomenklatura over the state.

In line with the common optimist picture of egovernance, it is pointed out that in India, compared to the previous citizen–administration relations characterized by bureaucratic rigidity, long delays, unnecessary complexity and public suffering, this relationship under e-governance is now characterized by higher speed, greater access, less cost and less public harassment [Pardo, 2000; Budhiraja, 2001].

An OECD study of egovernance, based on a series of interviews with information specialists, public officials and the policymaking community in eight post-industrial societies in 1996-7, found that digital technologies like email have had greater impact in the dissemination of information to senior

NY: Routledge; Christopher Weare, J. Musso, M.L. Hale. 1999. ‘Electronic democracy and the diffusion of municipal web pages in California.’ Administration & Society. 31(1): 3-27; Chris C. Demchak, Christian Friis, Todd M. La Porte. 1998. ‘Configuring Public Agencies in Cyberspace: Openness and Effectiveness.’ www.cyprg.arizona.edu/Tilburg98F.htm; Jerry Mechling. 1994. ‘A Customer service manifesto: using IT to improve government services.’ Government Technology. 1:S27-33; Dan Jellinek. 2000. ‘E-Government – Reality or Hype?’ iMP: Information Impacts Magazine. October. www.cisp.org/imp * Pippa Norris. 1999. Critical Citizens: Global Support for Democratic Governance. Oxford: Oxford University Press; C. Thomas. 1998. ‘Maintaining and restoring public trust in government agencies and their employees.’ Administration and Society. 30: 166-193.

15


decision-makers and policy elites, although even here traditional channels remained most popular, including press releases, official Gazettes and face-to-face meetings.

The studies conducted by the Center for Electronic Governance at Indian Institute of Management, Ahmedabad, indicate that the governments are enthusiastic in adopting egovernance. There are many examples of egovernance projects, which have won international and national awards. However any government services need to be re-engineered to benefit from the emerging ICTs. There is an inherent distrust in citizens on the service delivery mechanisms. This image needs to be improved with confidence building measures.

Shackleton et al [2004] examined the current status of Australian local government electronic service delivery and explored the appropriateness of current e-Business maturity models for evaluating the progress local governments are making towards electronic service delivery. Their research involved an evaluation of local government websites and a detailed case study of one local council. The results indicate that apart from Web based information provision, little progress has been made in the transition to electronic service delivery in most areas of local government.

Wadia [2000] mentioned that in India, egovernance creates an avenue for its citizens to communicate with top political leaders and local ministers through such tools as video-conferencing, online grievance channels and complaint cells. In her comparative studies based on the Inter-Parliamentary Union list, Norris [2001] observes that there are 98 countries in which the national parliaments have their own websites; of these the most comprehensive ones are from Scandinavia, Western Europe and North America.

Moreover she found that among the developing countries, the website of the Indian Parliament (alfa.nic.in) is quite comprehensive. It encompasses a list of basic information regarding the House of People (Lok Sabha) and the Council of States (Rajya Sabha). The menu includes such items as parliamentary activities, parliamentary committees, budget matters, national constitution, legislative acts, Prime Minister’s office, web addresses of all ministries and states, bulletins and publications, economic surveys, citizen services, and profiles and speeches of parliamentary members. It also provides an option for citizens to send feedback and suggestions through email. The Prime Minister’s Office also has a website, which provides information regarding his policy initiatives maintains an option for surveying opinion regarding current political issues and offers opportunities for the public to send queries and comments. These online sources of information and avenues for public expression are supposed to be more conducive to a stronger relationship between citizens and politicians.

Heeks [1998b] found that out of 400–500 software export firms in India, the top 20 firms were responsible for 70 percent of all exports. Geographically, most of the 558 Indian software company headquarters are located only in few large cities: 152 in Bangalore, 122 in Mumbai, 93 in Chennai, 86 in Delhi, 34 in Hyderabad, 27 in Calcutta, 22 in Pune, and remaining 22 in all other cities. These unequal structures of IT resulting from policies pursued under e-governance, thus, imply greater economic and geographical divides in India.

3. OVERVIEW OF THE MAURITIAN CASE

Mauritius has exciting plans for becoming an information society or, in its own words, a Cyber Island*. The concept of building an information economy goes back to the early 1990s†. However it is only recently that top-level commitment backed by funding for specific ICT projects has given Mauritius a new momentum. This is manifested in the government’s intention to make Information and Communication Technology the fifth pillar of the Mauritian economy alongside sugar, Export Processing Zones, financial services and

* “Apart from infrastructure development, which is well underway, we need to focus on three other critical factors for transforming Mauritius into a CyberIsland. These are human resource development, telecoms connectivity and access to computers at home.” http://www.cdacindia.com/html/pdf/pmspeech.pdf. † For example a National Seminar on Information Technology was held in December 1993 and a joint National Computer Board - World Bank report entitled Information Technology and the Competitive Edge was issued in June 1995. See http://www4.worldbank.org/afr/poverty/pdf/docnav/02375.pdf. [Accessed 21 July 2004]. A National Information Technology Strategy Plan was issued in 1997. See http://ncb.intnet.mu/ncb/nitsp/.

16


tourism. Thus, a key ingredient for the development of an information society -high-level government recognition and support -is present. The Government of Mauritius realizes that the historical perception of public service is characterized by queues and slow procedures. It is keen to overcome that stereotype by using ICTs to offer “efficient, effective and citizen-focused public services 24 hours a day, 7 days a week.”* One step in that direction beginning in 1996, has been putting all ministries online with web sites containing information about their work, including legal texts, publications, events, services available to the public and contact details. A growing number of government departments provide downloadable forms online, adding to convenience for citizens. Over 100 forms were available in early 2003 including applications for passports and driving licenses and tax and business registration forms. Public administration web sites — including the 26 ministries and over 80 governmental bodies — are linked through the Government of Mauritius web portal [www.gov.mu].

Several interactive government services are available mainly targeted at the business community. These include TradeNet, introduced back in 1994. The system allows import and export businesses to submit declarations and other documents electronically to the Customs and Excise Department. Another example is the Contributions Network Project, available since January 2002, allowing companies to file taxes online. Mauritius is now moving towards a higher level of achieving its vision of around the clock availability of all government services for citizens through several egovernment initiatives and projects. This involves making the transition from static web sites to fully integrated transactional services. EGovernment would be accessible from a variety of access points including traditional counter services, as well as the Internet, kiosks and call centres. Perhaps the most visible government initiative is the Cyber City project. The project is expected to have a spill over effect and spread ICT throughout Mauritius, from the Cyber Tower, to the Cyber City and finally to the Cyber Island [http://e-cybercity.mu].

The E-government Task Force, chaired by the Minister of Information Technology and Telecommunications, has been charged with overseeing implementation of the eGovernment programme. A concept paper lays out the vision, objectives and benefits of e-government and sets the deadline of 2005 for having all government services online. Key players to implement the programme are the Central Informatics Bureau and Central Information Systems Division working with a Chief Information Officer in each ministry.

The eGovernment programme consists of a number of projects of which three main ones are currently being implemented: Government Online Centre (GOC), Government Intranet System and Egovernment: Online Delivery Services. The heart of the system is the GOC. The total budget for implementing GOC is Rs 40 million for the next two years. A tender for Phase I was launched in January 2003. This phase will develop an E-Centre and Government Portal. Phase II of the project will focus on disaster recovery and Phase III will be a Government Call Centre to handle queries. Moreover, the National ICT Strategic Plan 2007-2011 of the government sets out the Government’s vision to make ICT the fifth pillar of the economy by increasing ICT contribution to GDP and building collaborative ventures in the field of ICT with countries of the region. This new plan also includes programme monitoring indicators and milestones that will ensure right track is followed in achieving ICT targets. To realise the above vision the Plan scales up Mauritius in terms of five strategic trust areas namely providing support to legal, institutional and infrastructural framework related to ICT, promoting e-business adoption, accelerating ICT adoption in society, transforming the island into an ICT expertise hub in the region to take up leadership roles and finally becoming an investment nucleus for ICT and thus emerging as a global point of reference for offshore services in the fields of ITS and ITes [http://ncb.intnet.mu/mitt/ministry/ICT/cpaper.htm].

4. METHODOLOGY

The core concept underlying all research is its methodology as it controls the study, dictates the acquisition of the data, and arranges them in logical relationships. For the purpose of this research, data was mainly collected through survey. Personal interviews were not used due to high costs involved and no response problems. This project uses semi-structured type of interviews conducted via the administration of questionnaires. Moreover to avoid misunderstandings and errors the questionnaires were administered

* EGovernment Task Force. Available at:http://ncb.intnet.mu/mitt/ministry/ICT/cpaper.htm.

17


personally and the respondents were assisted in ‘creole’, their native language. Different websites of eservices of government have also been used as secondary data. The eservice chosen for case study analysis is the: online application for a Learner's Driving License. To apply for a Learner’s License the registered applicant needs to fill in the online Learner's Driving License form, which is then electronically submitted to the Traffic Branch Section of the Police Force for processing.The targeted population for this project consists of adults, that is, above 18 years of age but below 60 years, as they represent the people who will normally apply for driving license. Out of a targeted population of approximately 745,653 (based on 2000 Census from Central Statistical Office), 200 adults were randomly chosen to fill the questionnaires. The questionnaires were distributed on a pro-rata basis across the different districts and equally between males and females given that each gender group represents around 50% of the targeted population. The questionnaires were designed in such a way that it can answer the questions set out under the statement of problem. Questions were designed to gauge people’s views on how the application for learner’s license online has impacted on service delivery in terms of the quality effects [Yang, 2001], process effects [Ciborra, 2005] and operation effects [Riley, 2003 and Saxena, 2005]. When the draft questionnaires were ready, they were pre-tested through consultation with members of the public. The questionnaires had to be pilot-tested to check the grammar, wording, sequencing, layout and estimated rate of response. Tests were carried out with a group of 5 persons form the public.

5. RESULT FINDINGS

About quality, the respondents describe that the supporting and interface cover the service very well and is quite fast, but however have a medium level as the information they provide is static. Alongside the online service does not cover all the physical service. So if someone wants to know some information about specific aspects, it is necessary to go to the office. Moreover they said that an information contact number is available where one can ask what one wants but if someone needs to contact personally with an individual involved in the register one will have to go to the physical place. This variable goes against the expectations of Yang [2001] as easy-to-follow catalogues, site navigability and concise and understandable contents.

According to the responses obtained one of the most critical reasons for e-governance being less effective is the problem of citizens’ access to the available information sources such as the internet. Infrastructure such as availability of computers, electricity and telephone is not a problem in the islands. The major problem is the low rate of internet connectivity, though price is not a major hindrance in this context. Another barrier may be the lack of trust in terms of security on the part of users. However in Mauritius there are laws against electronic fraud, such as the Computer Misuse and Cybercrime Act 2003 and the Data Protection Act 2004 so that this fear may be reduced by a change in attitude and culture.

In addition, the dominance of English on the internet constrains the access of non- English-speaking population. Thus they recommend that in order to strengthen the citizen–administration relationship, many state governments have taken other measures such as the introduction of local languages onto their websites Alongside they argue that a more mature site would enable a user to seek support for a service product or service without having to wait until the office concerned is opened and this is not the case with this eservice. Services such as the ability to track the progress of application, a common over-the-counter query, were not available on the site. In this regard, the Internet was used conservatively, as predicted, to replicate existing channels for the publication and distribution of official documents like reports, providing information through different channels, rather than to ‘reinvent government’, to rethink the nature of the relationship between departments and the public, or to open bureaucratic organizations to interactivity with customer-clients.

They stated that, not only clickable feedback and email options are needed but it also important to assess the actual quantity of feedback and suggestions and the frequency at which public officials genuinely respond to them. An important consideration in this regard, according to them, is how relevant the online discussion items and information sources are to the needs and interests of various segments of the population in Mauritius.

Concerning reliability in punctual delivery of the service, the majority of respondents identify it as being of low level. About responsiveness to citizens within a promised time frame, only a minority of the respondents agree that the eservice has catered for this aspect. With regards to ease of access to the

18


representatives of the service, most respondents identify it as low level, since one has to go physically to have access directly to representatives. Related to ICT process, the respondents say that ICT itself and customized services is totally reviewed I this site because users can check in a personalized way with a code and password. These variables correspond to Surjadjaja et al. [2003] as a development of a self-service experience as well as the functionality of the ICT as a way to offer personalized services. Moreover based on the responses, in terms of E-transactions the site is not clear because one can fill in directly in the site but will have to submit the papers and certificates needed in the physical office. The respondents do not consider the possibility of e-transaction suggested by Ciborra [2005] as one of the most interesting in this eservice.

According to operations, all aspects as e-consultation, closer Government and modernization service are more or less enclosed in this eservice. This is connected to Saxena [2005] variables as communication between public servants and citizenry.

6. CONCLUSION

The purpose of this study was to contribute to a better understanding of the effect of eGovernance via the introduction of a specific eservice (application for learner’s licence) on service delivery in the Mauritian Public Sector. Through the use of a case study we have tried to yield sufficient information in order to provide answers to the research questions posed. The research findings indicate that eGovernance has improved service delivery of that specific eservice in terms of clearer information, better quality, modernized and personalized service and speedy process. However the eservice still has certain weaknesses in terms of double processes (physical and online), wrong communication and lack of options for feedback. It must be noted that this study has been limited to investigating the Mauritian Government in the fast movement of eGovernment and hence no real generalisable conclusions can be drawn from that specific setting.

7. REFERENCES

BUDHIRAJA, R. 2001. Electronic Governance — A Key Issue in the 21st Century. Ministry of Information Technology, India. [http://egov.mit.gov.in].

CIBORRA, C. 2005. Interpreting eGovernment and development: Efficiency, Tranparency, or Governance at a distance? Journal of Information Technology and People. Emerald Group Publishing, 18(3).

DALY, J.A. 2000. Will the Internet Promote Democracy?’, iMP Magazine, September. [http://www.cisp.org/imp/september_2000/daly/09_00daly.htm].

DAVENPORT, T., AND SHORT, J. 1990. The new industrial engineering: Information technology and business process redesign. Sloan Management Review. 31(4) 11-26.

FARNHAM, D., AND HORTON, S. 1996. Managing the New Public Services, London, MACMILLAN Press.

DUNSIRE, A. 1995. Administrative Theory in the 1980’s, Public Administration, 73, 1-15. GARSON, G,D. 2000. Hand Book Of Public Information Systems, New York, Marcel Dekker, Inc. GARSON, G,D.1999. Information Technology and Computer Applications for Public Administration,

Issues and Trends, London, Idea Group Publications. GHERE, R,K. and YOUNG, B,A. 1998. The Cyber-management Environment: Where Technology and

Ingenuity Meet Public Purpose and Accountability. Public Administration and Management: An Interactive Journal 3(1). [http://www.pamij.com/gypaper.html].

HEEKS, R. 1998a. Information Age Reform of the Public Sector: The Potential and Problems of IT for India. Working Paper No. 6. Manchester: Institute for Development Policy and Management, University of Manchester.

HEEKS, R. 1998b. The Uneven Profile of Indian Software Exports. Development Informatics. Working Paper Series, Working Paper No. 3. Manchester: Institute for Development Policy and Management, University of Manchester.

HEEKS, R. 1999. Information and Communication Technologies, Poverty and Development. Development Informatics. Working Paper Series, Working Paper No. 5. Manchester: Institute for Development Policy and Management, University of Manchester.

19


HEEKS, R. 2001a. Understanding e-Governance for Development. i-Government Working Paper Series, No. 11. Manchester: Institute for Development Policy and Management, University of Manchester.

HEEKS, R. 2001b. Building e-Governance for Development: A Framework for National and Donor Action. i-Government Working Paper Series, No. 12. Manchester: Institute for Development Policy and Management, University of Manchester.

KLING, R. 2000. Learning from Social Informatics: Information and Communication Technologies in Human Contexts. Center for Social Informatics. Indiana University.

LEVINE, P. 2001. The Internet and Civil Society: Dangers and Opportunities. iMP Magazine, May. [http://www.cisp.org/imp/may_2001/05_01levine.htm]

MASSEY, A. 1993. Managing the Public Sector. Aldershot: Edward Elgar. Hughes, O.1994. Public Management and Administration. New York: St. Martin’s Press.

National Research Council. 2002. Computer Science and Telecommunications Board Information Technology Research, Innovation and E-Government. National Academy of Sciences, Washington DC.

NORRIS, P. 2001. Digital Divide? Civic Engagement, Information Poverty and the Internet in Democratic Societies. Cambridge: Cambridge University Press.

PARDO, T,A. 2000. Realizing the Promise of Digital Government: It’s More than Building a Web Site. IMP Magazine, October. [http://www.cisp.org/imp/ october_2000/10_00pardo.htm]

PC World 2000. Cyber States in India and E-governance. [http://www.pcwindia.com/2000jun/cs.htm]. SCHWARE, R. 2000. Information Technology and Public Sector Management in Developing

Countries:Present Status and Future Prospects. Indian Journal of Public Administration 46(3), 411–16.

SINGH, S,H. 2000. Ways and Means of Bridging the Gap between Developed and Developing Countries. Paper presented at the Panel on Information Technology and Public Administration at United Nations, New York, 26 September.

UNDP 1999. Human Development Report 1999. New York: Oxford University Press. UNDP 2000. Human Development Report 2000. New York: Oxford University Press. WACHBROIT, R. 2001. Reliance and Reliability: The Problem of Information on the Internet.

iMPMagazine, May. [http://www.cisp.org/imp/may_2001/ 05_01wachbroit.htm]. WADIA, J. 2000. Welcome to Digital Democracy. Times Computing, 22 November.

[http://www.timescomputing.com/20001122/nws1.html]. YIN, R, K. 1994. Case Study Research. Sage Thousand Oaks, CA.

20


Correlates of Computer Attitude among Secondary School Students in Lagos State, Nigeria

ADEBOWALE, O.F, ADEDIWURA, A.A & Bada, T. A* Obafemi Awolowo University, Ile-Ife. Nigeria Abstract Research have been conducted to study students’ attitude towards the computer [computer attitudes], their computer self efficacy and computer anxiety separately. This study was specifically targeted at determining if socio-demographic variables like gender, age and field of study had any effect on these computer parameters among secondary school students. It also explored the possibility of been able to predict students’ computer characteristics from computer efficacy, computer anxiety and demographic variables. 600 students were selected by proportionate sampling from the Senior Secondary class III [SS III] of six secondary schools equipped with 40 micro-computer-fitted laboratories by the Nigerian [Lagos state] government. The instrument for the study consisted of two types of questionnaires, one titled “Questionnaire on the computer attitude” was used to obtain a measure of students’ computer attitude while another titled “Questionnaire on students computer self efficacy and computer anxiety” was used to measure their computer self efficacy and computer anxiety. The questionnaires were administered by their ICT teachers under the supervision of the researchers. Data analyses were by using t-test, ANOVA, Post-hoc tests and multiple regression. The results showed that gender had no significant influence on any of the three parameters but age seems to affect computer attitude and computer anxiety. Students in the vocational and commercial fields of study had better attitude towards the computer than those in the sciences and arts. In terms of predicting students computer attitudes, fields of study, computer self efficacy, gender and very low levels of computer anxiety were found to be the significant predictors of computer attitude. Categories and Subject Descriptors: J [Computer Applications] J.4 Social and Behavioral Sciences – Psychology. General Terms: Computer attitude, Computer self efficacy, computer anxiety. Additional Keywords and Phrases: Computer characteristics, socio-demographic variables, field of studies, students’ attitude toward the computer IJCIR Reference Format: Adebowale, O.F, Adediwura, A.A., Bada, T. A., Correlates of Computer Attitude among Secondary School Students in Lagos State, Nigeria. International Journal of Computing and ICT Research, Vol. 3, No. 2, pp. 20 - 30. http://www.ijcir.org/volume3-number2/article3.pdf. 1. INTRODUCTION

* Author’s Address: Adebowale, O.F, Department of Educational Foundations and Counseling, Obafemi Olowo University, Ile-Ife. Nigeria, [email protected], [email protected]; Adediwura, A.A., Department of Educational Foundations and Counseling, Obafemi Awolowo University, Ile-Ife Nigeria, [email protected]; Bada, T. A., Department of Educational Technology, Obafemi Awolowo University, Ile-Ife. Nigeria, [email protected]

"Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than IJCIR must be honored. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee." © International Journal of Computing and ICT Research 2008. International Journal of Computing and ICT Research, ISSN 1818-1139 (Print), ISSN 1996-1065 (Online), Vol.3, No.2 pp. 20-30, December 2008.

21


We live in a postmodern society, where information is considered to be an extremely valuable commodity. Those who control important information, or who simply know how to access and use it, are the key players in the information-based economy [Simmons, 2009]. He further argued that, computer literacy and the skills that can be built there from are essential to one’s effectiveness in modern societies, not just in our working lives, but in the way we learn, manage our finances, and improve our standard of living.

When it comes to teaching and learning, Computers can be an incredible tool, especially when the learners have access to data stored on CD-ROMs or the Internet. They can use a PC to access vast knowledge bases on almost any topic, search archives of information dating back decades, ask questions online and even take online courses [ERS, Undated]. So it is important to have a basic understanding of computer technology, regardless of one’s career choice or aspiration.

Researchers have proposed that positive attitudes toward computers, high computer self-efficacy and lower computer anxiety levels could be important factors in helping people learn computer skills and use computers [e.g., Busch, 1995 in Sam, Othman and Nordin, 2005]. Woodrow [1991] claimed that students’ attitudes toward computers were critical issues in computer courses and computer-based curricula. Sam, Othman and Nordin [2005] concluded that monitoring the user’s attitudes toward computers should be a continuous process if the computer is to be used as a teaching and learning tool. Other attributes, such as gender and age [Morris, 1988-1989] and computer anxiety [Paxton & Turner, 1984] were also shown to be related to attitudes toward computers.

2. COMPUTER ATTITUDE

In general, attitudes can be defined as “a learned predisposition to respond in a consistently favorable or unfavorable manner with respect to a given object” [Fishbein and Ajzen, 1975]. They are relatively less stable than personality traits and can be changed both across time and across situations in virtue of individual’s interaction with the environment [Robinson, Simpson, Huefner, and Hunt, 1991]. Since attitudes are learnt, they are mouldable i.e. they change with experience of the stimulus objects and with social rules or institutions [Binder and Niederle, 2007].

According to Whitrow [1999] computer-related attitudes influence students’ desire to use computers, their desire to enroll in computer-related subjects and courses, and their choice of career path. Students’ computer-related attitudes are also directly related to their prior experiences and use of computers [Levine & Donitsa-Schmidt, 1997].

Different researches have been conducted on how users’ attitudes toward computers [herein referred to as computer attitudes] influence the future use of and behaviour toward computers [e.g. Fann, Lynch & Murranka, 1989; Woodrow, 1991; Levine & Donitsa-Schmidt, 1997]; the use of computers in optional circumstances [Fann et al., 1989], acceptance of computers [Selwyn, 1997] as well as future subject enrolment at school and the attendant selected career path [Busch, 1995; Levine & Donitsa-Schmidt, 1997]. Consequently, Nash and Moroz [1997] supported the view that evaluation of computer attitudes is an important technique in response to the trend of computers becoming more centralized in education through integration. This study is focused on factors which may be responsible for certain attitudes the student demonstrates towards the use of computer and which may affect their future interest and possible choice of computer and related vocations. 2.1 Computer Self Efficacy Kinzie, Delcourt, and Powers [1994] defined self-efficacy as an individual’s belief in his or her ability, which may impact the performance of tasks:

“Self-efficacy reflects an individual’s belief in his/her ability to perform the behavior required to produce specific outcome and it’s thought to directly impact the choice to engage in a task, as well as the effort that will be expended and the persistence that will be exhibited.” [p. 747]

Self-efficacy has been shown to influence choice of whether to engage in a task, the effort expended in performing it, and the persistence shown in accomplishing it [Bouffard-Bouchard, 1990]. Also Brown [2008] cautioned that self efficacy is not the same as actual knowledge of a task or with self esteem, which actually refers more to feelings of self-worth, but one which is situational and which highly influences people’s decision, goals, the amount if time they persevere through obstacles and difficulties.

22


Individuals who perceive themselves capable of performing certain tasks or activities are defined as high in self-efficacy and are more likely to attempt and execute these tasks and activities. People who perceive themselves as less capable are less likely to attempt and execute these tasks and activities, and are accordingly defined as lower in self-efficacy [Barling & Beattie, 1983; Bandura, Adams, & Beyer, 1977 in Karsten and Roth, 1998].

Self-efficacy has been suggested to be an important factor in the acquisition of computing skills [Miura,1987]. Computer self-efficacy is a belief of one’s capability to use the computer [Compeau & Higgins, 1995] such that participants with little belief in their ability to use computers might perform more poorly on computer-based tasks whereas better computer self-efficacy could increase persistence and success in studying computing [Sam, Othman and Nordin, 2005] and carrying out computer-based tasks. Khorrami-Arani [2001] highlighted several researches which demonstrated the impact that computer self-efficacy may have on increasing performance and the technological innovation of employees, reducing computer induced anxiety, and promoting higher occupational positions.

Research has focused on the relationship of a number of individual and situational variables to computer self efficacy. The relationship of gender to computer self efficacy has been of regular interest, since gender differences in self-efficacy have been investigated [Murphy, Coover and Owen, 1989], but findings have been mixed, for instance, Harrison & Ranier [1992] found that males demonstrated higher computer self efficacy than females but Smith [1994] found no gender differences on a measure of computer task self-efficacy among university students enrolled in an introductory computer science course, also in a study examining graduate students, adult vocational students, and professionals [nurses] in three different computer training settings [Murphy et al., 1989]. Also the relationship of computer experience to computer self efficacy has been investigated but no research has been found to determine the relationship between computer self efficacy, computer anxiety and students’ attitude to the use of computers. 2.2 Computer Anxiety According to Dukes, Discenza and Couger [1989] studies conducted by Weinberg [1980] and Weinberg, English and Mond [1981] were the earliest to report the existence of computer anxiety. They described it as a response to interaction or anticipation of interaction with automated data or information processing system. Further studies have given more explicit definitions. For instance Computer anxiety has been defined as a fear of computers when using one, or fearing the possibility of using a computer [Chua, Chen, & Wong, 1999]. Thatcher and Perrewe [2002] saw computer anxiety as individuals' judgment of their capabilities to use computers in diverse situations.

Computer anxiety is characterized as an affective response, an emotional fear of potential negative outcomes; it may include worries about embarrassment, looking foolish or even damaging computer equipment [McInerney, McInerney & Sinclair, 1994 in Phelps and Ellis 2002]. Maurer and Simonson [1984:6] clarified the difference between computer anxiety and the rational fears related to computer utilization such as in job placement, increased exposure to radiation from terminal screens, and concluded that computer anxiety relates to feelings of impending doom or sure calamity because of contact with computer. Consequently they identified four behavioural indicators of computer anxiety:

1. Avoidance of computers and the general areas where computers are located. 2. Excessive caution with computers. 3. Negative remarks about computers and 4. Attempts to cut short the necessary use of computers. From an information processing perspectives, the negative feelings associated with high anxiety

detract cognitive resources from task performance [Kanfer & Heggestad, 1997]. Thus the performance of participants with higher computer anxiety might be poorer than those with little or no computer anxiety. Raub [1981] found five contributing factors to computer anxiety to include gender, computer experience, college major, math anxiety, and trait anxiety. He also found that computer attitudes are gender-specific and culturally-learned. On the other hand, Thatcher and Perrewe [2002] posited that individuals who have more confidence in their capabilities tend to demonstrate lower levels of computer anxiety. From the foregoing, investigating computer anxiety as a possible correlate of computer attitude may not be out of place. 2.3 Purpose of the research In reference to results of an earlier study carried out by Adebowale and Adewale [in press] and considering the importance of students’ computer attitude to their future use and possibility of taking up future computing careers, the researchers saw an important need to determine the factors which may influence the

23


development of a healthy computer attitude in learners. As a result, the study looked at two research objectives. It sought to determine if socio-demographic variables like gender, age or field of study had any effect on computer attitudes, self efficacy and computer anxiety of learners. It also sought to determine which of the earlier stated factors are significant correlates of computer attitude among students. Specifically, this research investigated the following research hypotheses. Research Hypothesis One: The students’ gender has no significant influence on computer attitude, self efficacy and computer anxiety. Research Hypothesis Two: There is no significant difference in the computer attitude, self efficacy and computer anxiety of the secondary school students under study on the basis of age groupings? Research hypothesis Three: There is no significant difference in the computer attitude, self efficacy and computer anxiety of the secondary school students under study on the basis of students’ fields of study. Research hypothesis Three: The identified factors are not significant predictors of computer attitude among the students under study.

3. METHODOLOGY

The research design used for the study is a descriptive survey design. Six hundred [600] Senior Secondary school class III [SS III] students were selected by proportionate random sampling from the three zones which makes up the Educational District I [ED I] of Lagos state in the first term of the 2007/2008 academic session. The district consists of three zones [each coinciding with a local government area] of Agege, Ifako/Ijaye and Alimosho. Two schools equipped with forty microcomputer-fitted laboratories each by the state government were selected in each of these zones and one hundred [100] students were randomly selected from the SS III classes of each of the six schools to participate in the study.

The instruments used were two types of questionnaires. The first one was adopted from Adebowale and Adewale [in press], consisting of three sections which together were used to produce a measure of the students’ attitude to computer. Section A consists of 12 items, section B has 13 items while section C contain 8 items. Some of the items were adapted from the Computer Attitude Scale developed by Nickell & Pinto [1986] while others were derived from other Computer attitude questionnaires developed by the Texas Center for Educational Technology [1998], Internet Education Research Group [undated] and the Institute for the Integration of Technology into Teaching and Learning [2003]. The respondents were required to provide response in five categories corresponding to their level of agreement with the statements given as “Strongly agree” “Agree” “Undecided” “Disagree” “Strongly disagree”. Copies of the questionnaire were earlier circulated to 42 students who did not eventually participate in the study. Their responses were used to obtain validity and reliability information. It also necessitated deleting three items from section A, two from section B and four from section C. The final version of the instrument gave a Cronbach’s Alpha estimate of 0.81 and split-half value of 0.79 showing that the instrument can be said to be very reliable for the type of study for which it is designed.

The second Questionnaire titled “computer anxiety and self efficacy measures of secondary school students” was used to measure the computer anxiety and self efficacy of the students under study. It consists of three sections – section A which was used to collect socio-demographic information such as gender, age, field of study, name of school and so forth. Section B was used to collect information on the respondents’ self efficacy and contained items adapted from Khorrami-Arani [2001] and Karsten & Roth [1998]. Section C contain items which measured the computer anxiety of the respondents, the items were adapted from the Computer Anxiety Scale from the University of North Carolina at Charlotte [see http://www.psych.uncc.edu/pagoolka/ComputerAnxiety.html] and the Computer Anxiety Rating Scale of the University of Southern Maine [see http://www.usm.maine.edu/com/carssc~1.pdf]. The three sections were compiled into a single composite and assembled into the instrument used for this study. For this section, the respondents were required to provide responses in five categories corresponding to their level of agreement with the statements given as “Strongly agree” “Agree” “Undecided” “Disagree” “Strongly disagree”. Copies of the questionnaire were also circulated to the 42 students mentioned earlier. Their responses were also used to obtain validity and reliability information. The final version of the instrument gave a Cronbach’s Alpha estimate of 0.77 and split-half value of 0.84 showing that the instrument can be said to be reliable for this type of study.

The instrument was administered on the respondents by their ICT teachers under close supervision of the researchers. Out of the 600 pieces of the questionnaire circulated only 540 [90% return rate] could be

24


used for the study; others were either not returned or not properly filled. Data analysis was by using, t-test, ANOVA and post-hoc analysis provided in the SPSS 14 Software. The results were as presented below.

4. RESULTS

Research Hypothesis One: Students’ gender has no significant influence on computer attitude, self efficacy and computer anxiety among the secondary school students under study?

To test this hypothesis, the respondents’ scores on their attitudes, computer self efficacy and computer anxiety were independently subjected to tests of difference in means using gender as the basis for differences. The Respondents scores on attitude toward the computer ranged from 90 to 170, for computer self efficacy their score ranged from 6 to 85 and for computer anxiety, the scores ranged from 4 to 80. Table 1: Test of difference in male and female students’ computer characteristics Computer characteristics Respondents’ sex N Mean Std. Deviation T Sig. [2-tailed] respondent computer attitude Male 124 131.5242 13.50186

-.869 .386 Female 94 133.0426 11.74085

Computer Self-Efficacy Male 123 66.4634 11.29071-.230 .818

Female 92 66.8261 11.60012computer anxiety Male 116 40.2414 15.14086

-.559 .577 Female 89 41.4494 15.61396

Table 1 above shows that in the three cases the null hypothesis cannot be rejected as generally, there is no gender difference in the computer attitude of the students under study [t = -0.869, p >.05]. Also, male and female respondents demonstrated no significant difference in the computer self efficacy [t = -0.230, p > .05], and computer anxiety [t = -0.559, p > .05] experience by them. Research Hypothesis 2: There is no significant difference in the computer attitude, self efficacy and computer anxiety of the secondary school students under study on the basis of age groupings?

To address this research hypothesis, the responses of the students on the basis of their age groups [Pre-Adolescents – 10-12yrs, early adolescents – 13 – 15 yrs, mid adolescents – 16 to 18 yrs; late adolescents – 19yrs and above] were subjected to analysis of variance and the result was as shown in table 2 below. Table 2: Test of difference in students computer characteristics on the basis of their age grades

It can be observed from table 2 above that the null hypothesis cannot be accepted for computer attitude as the p-value did not attain or surpass the mandatory 0.05 threshold. Hence we can conclude that there is a significant difference in the students attitude to computer on the basis of age [F3,215 = 4.033, p <.05]. This is also the case for computer anxiety [F3,203 = 3.377, p <.05], but no significant difference was demonstrated by different age groups in computer self efficacy [F3,213 = 4.033, p <.05]. The researchers attempted to find

Sum of Squares Df Mean Square F Sig. respondent computer attitude Between Groups 1955.658 3 651.886

4.033 .008 Within Groups 34753.502 215 161.644Total 36709.160 218

Computer Self-Efficacy Between Groups 346.335 3 115.445.864 .460 Within Groups 28448.144 213 133.559

Total 28794.479 216 computer anxiety Between Groups 2205.670 3 735.223

3.377 .019 Within Groups 44192.070 203 217.695Total 46397.739 206

25


the direction of the difference observed in the computer characteristics of the students by conducting a multiple comparison test via Turkey HSD test. The result was shown in table 3 below. Table 3: Multiple comparison test of difference on the basis of respondents’ age groupings via TurkeyHSD

From Table 3, it can be observed that in terms of computer attitude, heavy difference was found between pre-adolescents and late adolescents were not found to be significant [mean difference = 22.5, p < .05]. This can be interpreted to mean that late adolescent students seem to possess better attitude to computer use and its other concerns. On the other hand the significant difference obtained in respondents computer anxiety on the basis of their age groupings could be seen to stem from the difference between Early-adolescents and Mid-adolescents [mean difference = 5.68 , p < .05]

Dependent Variable [I] age grade [J] age grade Mean Difference [I-J] Sig.

respondent computer attitude

Pre-Adolescents

Early Adolescents -2.58824 .954

Mid-Adolescents -6.81481 .517

Late Adolescents -25.50000 .043*

Early Adolescents

Pre-Adolescents 2.58824 .954


Late Adolescents -22.91176 .059

Mid-Adolescents


Early Adolescents 4.22658 .079


Late Adolescents

Pre-Adolescents 25.50000 .043*

Early Adolescents 22.91176 .059

Mid-Adolescents 18.68519 .170

computer anxiety

Pre-Adolescents




Early Adolescents


Mid-Adolescents -5.68354* .036*

Late Adolescents 1.64646 .999

Mid-Adolescents


Early Adolescents 5.68354* .036*

Late Adolescents 7.33000 .899

Late Adolescents




*. The mean difference is significant at the 0.05 level.

26


Research hypothesis 3: There is no significant difference in the computer attitude, self efficacy and computer anxiety of the secondary school students under study on the basis of students’ fields of study. To test this research hypothesis, the responses of the students on the basis of their fields of study [Science, Arts, Commercial and Vocational] were subjected to analysis of variance and the result was as shown in table 4 below. Table 4: Test of Difference in students’ computer characteristics on the basis field of study

Sum of Squares Df Mean Square F Sig.

respondent computer attitude Between Groups 5728.271 3 1909.424 14.546 .000Within Groups 28353.165 216 131.265

Total 34081.436 219

Computer Self-Efficacy Between Groups 1172.452 3 390.817 3.185 .025

Within Groups 26262.979 214 122.724

Total 27435.431 217 computer anxiety Between Groups 1229.179 3 409.726 1.902 .130

Within Groups 43723.903 203 215.389Total 44953.082 206

On the basis of field of study, table 4 shows that significant difference was obtained in computer attitude of the students [[F3,216 = 14.546, p <.05]. as well as their computer self efficacy [F3,214 = 3.185, p <.05]. However, field of study was found to have no significant influence on computer anxiety [F3,203 = 1.902, p >.05]. The researchers also determine the direction of the differences in computer attitudes and computer self-efficacy of the respondents belonging to different fields of study via a multiple comparison analysis and the result was as shown below.

Table 5: Multiple comparison test of difference on the basis of respondents’ age groupings Dependent Variable [I] field of study [J] field of study Mean Difference [I-J] Sig.

respondent computer attitude Science Arts -5.53560 .070 Commercial -11.77336* .000 Vocational -30.62791* .002

Arts

Science 5.53560 .070 Commercial -6.23776* .003 Vocational -25.09231* .014

Commercial

Science 11.77336* .000 Arts 6.23776* .003

Vocational -18.85455 .100 Vocational Science 30.62791* .002

Arts 25.09231* .014 Commercial 18.85455 .100

Computer Self-Efficacy Science Arts -4.10879 .243 Commercial -4.30865 .143 Vocational -20.28571 .048

Arts

Science 4.10879 .243 Commercial -.19986 .999 Vocational -16.17692 .179

27


Commercial

Science 4.30865 .143 Arts .19986 .999

Vocational -15.97706 .183 Vocational Science 20.28571 .048

Arts 16.17692 .179 Commercial 15.97706 .183

*. The mean difference is significant at the 0.05 level. From Table 5 it can be seen that the largest significant difference observed in the respondents’ computer attitude was between science and vocational students [mean difference = 30.63, p = 0.002] in favour of those in vocational field, closely followed by the difference between arts and vocational [mean difference = 25.09, p = 0.014] and also between arts and commercial [mean difference = 6.24, p = 0.003]. In terms of computer self efficacy the only significant difference was spotted between science and vocational students [mean difference = 20.29, p = 0.048]. Research hypothesis 4: The identified factors are not significant predictors of computer attitude among the students under study. To test this hypothesis, the predictor variables are tested for their strength of predicting computer attitude among the students and the result was as presented below. Table 6: Multiple Regression of predictor variables in students’ computer attitude

Model

Unstandardized Coefficients

Standardized Coefficients

t Sig. R R

Square Adjusted R

Square B Std. Error Beta [Constant] 95.013 4.884 19.452 .000

.482a .232 .222

Computer Self-Efficacy .281 .050 .259 5.595 .000

computer anxiety .052 .040 .060 1.304 .003field of study 6.233 .781 .386 7.982 .000respondent sex 2.510 1.210 .099 2.074 .039respondent age -.134 .111 -.056 -1.202 .230a. Predictors: [Constant], respondent age, Computer Self-Efficacy, respondent sex, computer anxiety, field of study b. Dependent Variable: respondent computer attitude

From Table 6, it can be seen that all the variables except the respondents’ age seem to significantly predict computer attitude. The respondents’ field of seems to be the strongest factor which predicts his/her attitude to the computer with regression weight of 6.233 at p-value = 0.000 and contributing 38.6% of the observed variance, this was followed by sex [2.510, p-value = 0.039 contributing 9.9% of the observed variance]. Computer self efficacy was also a significant predictor [B = 0.281, p = 0.000 with 25.9% of the observed variance] as well as computer anxiety [ B = 0.052, p = 0.003 contributing only 6% of the observed variance]. Together the model they construct is a significant one with an R-value of 0.482 and accounting for 22.2% of any observed variance in the attitude of the students.

5. DISCUSSION

This study was designed to examine certain factors which may influence computer attitude in a developing economy like Nigeria. It is hoped that if these factors are known, facilitating environment could easily be organized for students to develop a healthy attitude towards computer use, promote persistence in studying computing and possibly encourage them to take up computer related vocations in future. Research

28


hypothesis one was posed to find out if the gender factor exercises any moderating influence on computer attitude, computer self efficacy and computer attitude. The results showed that gender has no significant influence on any of the three parameters. This led credence to earlier findings of Rohner and Simonson [1981] and Rosen, Sears and Weil [1987] as cited by Todman and Lawrenson[1992]. Kotrlik, and Smith [1988] also confirmed this findings.

In the second hypothesis, the study sought to find out differences in the computer attitude, computer self efficacy and computer anxiety of the students under study on the basis of their age groupings. The results indicated that age has nothing to do with computer self efficacy but could be fingered in the difference in their computer attitude and computer anxiety. This may be as a result of the fact that older students may have had more life and computing experience than the younger ones as the multiple comparison test indicated the difference between the old “Late adolescents” and the younger “Pre-Adolescents”. This is in consonance with the positive correlation found between age and computer attitude by Marshal and Bannon [1986] and Toddman and Lawrenson [1992] also quoted Rosen et al [1987] found a positive correlation between age and computer anxiety levels and attributed the inability of some researchers to find such a correlation to limited age range of the sample used in their study.

In Research hypothesis 3, it was found that the field of study the students pursue had a significant influence on their attitude towards the computer as well as on their computer self efficacy. It however had no significant effect on their computer anxiety. Students in the vocational fields of study seemed to possess better attitude towards the computer than other students, even better than what students in the commercial field demonstrated. Commercial students also demonstrated better attitude than students in the science and arts field. Perhaps students in the sciences and arts possess erroneous belief that they are suppose to give more attention to their school subjects rather learning, using or attempting to take up vocation in computing and its related fields.

After the influences of identified factors have been established, the researchers attempted to find out if the students’ computer attitude could be predicted from the identified factors and by so doing, make workable suggestions as to improving on the factors in order to enhance positive computer attitude in the learners. Consequently, a multiple regression test was conducted and the result indicated that learners’ field of study, computer self efficacy, gender and computer anxiety could be significant predictors of students’ computer attitude. Learners field of study seemed to be the strongest predictor, the next regression weight was that of computer self efficacy, all these are in agreement with the findings of Raub [1981]. Also gender seems to contribute to a significant prediction of computer attitude. A very low level of computer anxiety [very small but significant regression weight] seems to significantly predict computer attitude in agreement with Busch [1995]. 6. CONCLUSION The findings of this research has shown that effective management of socio-demographic factors [like gender and field of study], and personality variables [like computer self-efficacy and computer anxiety] could significantly predict how learners will relate to the computer, their persistence at studying computing and its allied courses as well as the development of interest in computer and computer related vocations.

Consequently, school counsellors and vocational guidance specialists have important roles to play in developing positive computer attitude in secondary school students by counselling them in gender relations to vocations and knowledge acquisition, usefulness of computers to students in all fields of study, counselling for confidence in handling computer and overcoming anxiety when using it. It is the view of the researchers that if these are properly managed, students attitude to computer, computing and computer vocations will be improved and many more will like to be involved in adopting computers and computing as a tool in the global march towards computerization and technological advancement.

However, it is suggested that the psychological basis of gender differences and contribution of these factors to computer attitude still require the attention of researchers as this will enable school counsellors to design appropriate guidance and counselling programmes which could be tailored towards improved attitude towards the computer, given the important roles computer and its applications play in the lives of man in the 21st century and beyond.

29


7. REFERENCES

ADEBOWALE, O.F AND ADEWALE, I.A. [in press] Computer Characteristics of Secondary School Students in Three Selected Local Government of Lagos State [Nigeria]: Implication for Global Computerization

BANDURA, A., ADAMS, N.E., & BEYER, J. [1977]. Cognitive processes mediating behavioral change. Journal of Personality and Social Psychology, 35 [3], 125-139.

BARLING, J. & BEATTIE, R. [1983]. Self-efficacy beliefs and sales performance. Journal of Organizational Behavior Management, 5, 41-51.

BINDER, M. AND NIEDERLE, U. [2007] Institutions as Determinants of Preference Change– A One Way Relation? Papers on Economic Evolution Number 0607 retrieved on 25th March 2009 from ftp://papers.econ.mpg.de/evo/discussionpapers/2006-07.pdf

BOUFFARD-BOUCHARD, T. [1990]. Influence of self-efficacy on performance in a cognitive task. The Journal of Social Psychology, 130, 353-363.

BROWN, J.H. [2008] Developing and Using a Computer Self-Efficacy Scale for Adults. Proceedings of the 24th Annual Conference on Distance Teaching and Learning retrieved on 10th April 2009 dfrom http://www.uwex.edu/disted/conference/Resource_library/proceedings/08_12667.pdf

BUSCH, T. [1995]. Gender differences in self-efficacy and attitudes toward computers. Journal of Educational Computing Research, 12, 147-158.

CHUA, S. L., CHEN, D., & WONG, A. F. L. [1999]. Computer anxiety and its correlates: A meta-analysis. Computers in Human Behavior, 15, 609-623.

COMPEAU, D. R., & HIGGINS, C. A. [1995]. Computer self-efficacy: Development of a measure and initial test. MIS Quarterly, 19, 189-211.

DUKES, R.L., DISCENZA, R. AND COUGER, J.D. [1989] Convergent Validity of Four Computer Anxiety Scales. Educational and Psychological Measurement 1989; 49; 195

ERS [Undated] Computer Literacy. Retrieved on 10th April 2009 from http://www.ers.sd83.bc.ca/assigns/it10B/fundamentals/computer_literacy.htm

FANN, G., LYNCH, D. & MURRANKA, P., 1989, ‘Integrating Technology: Attitudes as a Determinant of the Use of Microcomputers’, Journal of Educational Technology Systems, 17[4]: 307-317.

FISHBEIN, L AND AJZEN, I. [1975] Belief, Attitude, Intention and Behavior: An introduction to theory and research. Reading, MA; Addison-Wesley.

HARRISON, A.W. & RANIER, K. [1992]. The influence of individual differences on skill in end-user computing. Journal of Management Information Systems, Summer, 9 [1], 93-111.

KANFER, R., & HEGGESTAD, E. D. [1997]. Motivational traits and skills: A person-centered approach to work motivation. Research in Organizational Behavior, 19, 1-56.

KARSTEN, R & ROTH, R.M [1998] Computer Self-Efficacy: A Practical Indicator of Student Computer Competency in Introductory IS Courses Informing Science Volume 1 No 3

KINZIE, M. B., DELCOURT, M. A. B., & POWERS, S. M. [1994]. Computer technologies: Attitudes and self-efficacy across undergraduate disciplines. Research in Higher Education, 35, 745-768.

KHORRAMI-ARANI, O. [2001] Researching Computer self-efficacy International Education Journal Vol 2, No 4, 2001pg 17-25

KOTRLIK, J.W. AND SMITH, M.N. [1988]. Computer Anxiety Levels of Vocational Agriculture Teachers. Journal of Agricultural Education.............................

LEVINE, T. & DONITSA-SCHMIDT, S. [1997] Commitment to Learning: Effects of Computer Experience, Confidence and Attitudes. Journal of Educational Computing Research, 16[1]: 83-105.

MARSHAL, J., AND BANNON, S. [1986] Computer Attitudes and Computer Knowledge of Students and Educators, Association for Educational Data Systems [AEDS] Journal, Vol. 18, No. 4, [1986], 270-286

MAURER, M.W. & SIMONSON, M.R. [1984, January]. Development and Validation of a Measure of Computer Anxiety. Paper presented at the annual meeting of the Association for Educational Communications and Technology, Dallas, TX. [ERIC Documentation Reproduction Service No. ED 243 428]

MCINERNEY, V., MCINERNEY, D. M., & SINCLAIR, K. E. [1994]. Student teachers, computer anxiety and computer experience. Journal of Educational Computing Research, 11[1], 27-50.

30


MIURA, I. T. [1987]. The relationship of computer self-efficacy expectations to computer interest and course enrollment in college. Sex Roles, 16, 303-311.

MORRIS, D. C. [1988-1989]. A survey of age and attitudes toward computers. Journal of Educational Technology Systems, 17, 73-78.

MURPHY, C. A., COOVER, D., & OWEN, S. V. [1989]. Development and validation of the Computer Self-Efficacy Scale. Educational and Psychological Measurement, 49, 893-899.

NASH, J. & MOROZ, P., 1997, ‘An Examination of the Factor Structures of the Computer Attitude Scale’, Journal of Educational Computing Research, 17[4]: 341-356.

PAXTON, A. L., & TURNER, E. J. [1984]. The application of human factors to the needs of novice computer users. International Journal of Man-Machine Studies, 20, 137-156.

PHELPS, R & ELLIS, A 2002, 'Overcoming computer anxiety through reflection on attribution', in A Williamson, C Gunn, A Young & T Clear [eds], Winds of change in the sea of learning: charting the course of digital education: proceedings of the 19th Annual Conference of the Australasian Society for Computers in Learning in Tertiary Education [ASCILITE]: 8-11 December 2002, UNITEC Institute of Technology, Auckland, NZ, v. 2, pp. 515-524 retrieved from http://www.ascilite.org.au/conferences/auckland02/proceedings/papers/076.pdf on 28th March 2009

RAUB, A.C. [1981] Correlates of Computer Anxiety in College Students. A Ph. D Dissertation of the University of Pennsylvania. USA.retrieved on 12th April, 2009 from http://repository.upenn.edu/dissertations/AAI8208027/

ROBINSON, P.B., SIMPSON, D.V., HUEFNER, J.C. AND HUNT, H.K. [1991], “An attitude approach to the prediction of entrepreneurship”, in: Entrepreneurship.Theory & Practice, Summer, pp. 13-31.

ROHNER, D. J. & SIMONSON, M. R. [1981] Development of an index of computer anxiety. A paper presented at the Annual Convention of the Association of Educational Communications and Technology, Philadelphia, April [ERIC doc. no. 207 487] Retrieved on 6th April, 2009 from http://www.google.com.gh/search?hl=en&lr=&as_qdr=all&q=%22Development+of+an+index+of+computer+anxiety%22&btnG=Search

ROSEN, L. D., SEARS, D. C. & WEIL, M. M. [1987] Computerphobia, Behavior Research Methods, Instruments, & Computers, 19, pp. 167-179.

SELWYN, N. [1997] Students’ Attitudes Toward Computers: Validation of a Computer Attitude Scale for 16-19 Education. Computers and Education, 28[1]: 35-41.

SAM, H. K., OTHMAN, A. E. A., & NORDIN, Z. S. [2005]. Computer Self-Efficacy, Computer Anxiety,

and Attitudes toward the Internet: A Study among Undergraduates in Unimas. Educational Technology & Society, 8 [4], 205-219.

SIMMONS, V.G. [2009] Verizon Foundation Awards $8,000 Literacy Grant to Horry County Schools Adult Education. Retrieved on 10th April 2009 from http://www.edtheturtle.com/UserFiles/Servers/Server_1280957/File/Verizon_Release%5B1%5D.doc.

SMITH, J.M. [1994]. The effects of education on computer selfefficacy. Journal of Industrial Teacher Education, 31 [3], 51-65.

THATCHER, J.B. AND PERREWÉ, P.L. [2002] An Empirical Examination of Individual Traits as Antecedents to Computer Anxiety and Computer Self-Efficacy. MIS Quarterly, Vol. 26, No. 4 [Dec., 2002], pp. 381-396

TODMAN, J. AND LAWRENSON, H. [1992] Computer Anxiety in Primary Schoolchildren and University Students British Educational Research Journal, Vol. 18, No. 1, pp. 63-72

WEINBERG, S.B. [1980] Identification of computer anxiety. Connecticut Research Foundation 5170-000-0218-35-270

WEINBERG, S.B., ENGLISH, J.T. AND MOND, C.J. [1981] A stratagem for reduction of cyberphobia. Proceedings of the AAAS convention.

WHITROW, T.J. [1999]. Integrating Computers Across the Curriculum: Students' Computer-related Attitude Changes. Unpublished B Ed Honours Thesis, School of Education, Flinders University, Adelaide.

WOODROW, J. J. [1991]. A comparison of four computer attitudes scales. Journal of Educational Computing Research, 7, 165-187.

31


A Mobile Real Time Interactive Communication Assistant for Cerebral Palsy

LAU BEE THENG & HII KIING SHI* Swinburne University of Technology Sarawak Campus, Malaysia LOW TIONG KIE University Malaysia Sarawak, Malaysia Abstract

This research focuses on developing a mobile communication assistant for the disabled utilizing the biometrics information recognition in real time. We utilized their face features for our recognition. Our disabled community includes the palsies and autistics that could not make comprehensible speech due to disabilities or difficulties. However they have facial expression ability and incomprehensible speech that can be interpreted for their needs or requests. Their facial expressions may not be identical at all times for a specific need. Furthermore, each disabled is unique and uses a different way of expressions. Thus the communication assistant requires artificial intelligence in our proposed system. After a thorough research in face recognition and artificial intelligence, neural network coupled with Gabor feature extraction is found to outperform others. A neural network with Gabor filters is built to train the facial expression classifiers. Through our testing and evaluation with the volunteers, it has helped them expressing their needs with 85% successful recognition rates.

Keywords: Emotion communication, Gabor feature extraction IJCIR Reference Format: Lau Bee Theng, Low Tiong Kie and Hii Kiing Shi. A Mobile Real Time Interactive Communication Assistant for Cerebral Palsy. International Journal of Computing and ICT Research, Vol. 3, No. 2, pp. 31 -40. http://www.ijcir.org/volume3-number2/article4.pdf. 1. Introduction

Biometrics recognition is an automatic recognition of individuals based on their physiological or behavioral characteristics. Among the many body characteristics that have been used, facial expression is one of the most commonly used characteristics and has been studied across a number of research fields, e.g. computer vision, pattern recognition (Heisele et al. 2003, Zana and Roberto 2007, Zhao et al. 2003). Why do we choose facial expressions? According to psychologist, facial expression provides information about emotional states as well as cognitive activities (Ekman 2004). Emotions are revealed earlier through facial expression than people verbalize or even realize their emotional state (Zhao et al. 2003). Some psychologists also concluded that cognitive interpretations of emotions from facial expressions are innate and universal to all humans regardless of their culture (Zana and Roberto 2007, Zhao et al. 2003). Furthermore, facial expression recognition is regarded as a non-intrusive method that captures human

* Author’s Address: Lau Bee Theng & Hii Kiing Shi, Swinburne University of Technology Sarawak Campus, Malaysia; Low Tiong Kie, University Malaysia Sarawak, Malaysia "Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than IJCIR must be honored. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee." © International Journal of Computing and ICT Research 2009. International Journal of Computing and ICT Research, ISSN 1818-1139 (Print), ISSN 1996-1065 (Online), Vol.3, No.2 pp. 31- 40. December 2009.

32


subjects’ facial expressions through still images and/or video sequences in both controlled static environment and uncontrolled cluttered environment. This enlightens us with the use of biometric information to recognize the needs in real time and mobile basis for the disabled community.

2. COMMUNICATION ASSISTANT

We aim to develop a comprehensive communication assistant for the patient with cerebral palsy to assist all their need expressions. They have one common communication impediment: express needs in comprehensible speech. This is mainly due to some psychological and physical disabilities. However, they show expressions through their faces and non-verbal speeches when they have needs to be communicated to parents, teachers or caretakers. They have the abilities to show various facial expressions to signify different needs or incidents. Hence their expressions can be analyzed and interpreted to represent their speech. Naturally their facial expressions are not exactly identical at all times. Furthermore each of them is unique and has different ways to express a need. We aim to understand and capture their expressions, then form an expression pattern and assist them to express their desires.

For the initial development, we developed a communication assistant to understand a portion of their needs with an interactive, real time and mobile manner. We aim to incorporate the state-of-the-art of face recognition algorithms and hardware into our system (Figure 1). A disabled on the wheelchair or seat has his 2D frontal face image taken real time by the communication assistant installed in the laptop. The laptop is mounted on the wheelchair to go along with the disabled to give them mobility. The communication assistant performs both indoor and outdoor. The face image is captured automatically with a preset interval by parents, caretakers or teachers. The communication assistant can also serve as a monitoring system for them to track the activities of the disabled. Whenever a real time detected face is recognized, a voice message containing the needs is sent to the parents, caretakers or teachers through Bluetooth or wireless connections. Hence they can respond to the disabled immediately. This could help to save the time, increase the mobility and productivity of the parents, teachers or caretakers. When the caretakers or teachers change, any new one could substitute them easily with having any communication problems with the disabled. They could monitor more than one disabled at the same time through the communication assistant.

Figure 1 Prototyping of communication assistant, EmoCom

3. EXPRESSION RECOGNITION

The foremost task in our proposed communication assistant is the expression recognition. In order to have effective expression recognition for all the disabled, training data with the face images from disabled is essential. First of all, Eigen vector extraction is performed on the training images in the database, for each image of each disabled, a set of unique features is extracted. Each set of unique features is globalized into one feature vector to represent a disabled. Consequently, all the global feature vectors obtained from our human subjects are supplied as input to train a back propagation neural network. The rationale of utilizing neural network is our human subjects do not show exactly the same facial expressions at all times, variances occur within the same human subject’s expression for a particular need. However, the variations are within a range that can be trained and modeled through neural network. Hence a back propagation

33


neural network is fed with those variations for each disabled. The output is a set of trained facial expressions and non facial expressions for each disabled. Once a profile for a disable’s profile has been trained with sufficient images, then the communication assistant can be deployed to perform real time recognition. Once a detected frontal face is matched with a stored facial expression in our collection, speech may be expressed through our system to the caretakers, parents or teachers accordingly. For instance, if an expression is recognized as Child A’s tagged expression for hungriness in the our database, then a short message representing Child A will be sent to the respective recipient, “I am hungry, I want to eat now!”, hence the caretakers, parents or teachers can obtain the message clearly from a far, and attend to him with his food.

Image preprocessing First of all, the captured images from our disabled are transformed into gray scale images. The centers of two eyes on each gray scale image are used as the centers for rotation, translation, scaling and cropping. Each processed image has a size of 256×256 pixels. The preprocessed images are then subject to contrast/illumination and histogram equalization. Contrast is a measure of the human visual system sensitivity. The face recognition process in different lighting conditions with different illumination and contrast has different level of efficiency and psychologically meaningfulness. Hence, for our communication assistant application, all images are processed with same illumination and root mean square (RMS) contrast. The RMS contrast metric is equivalent to the standard deviation of luminance (Weyrauch

and Huang 2003). xi is a normalized gray-level value such that 0 < xi < 1 and −x is the mean normalized

gray level. With this normalization, images of different faces have the same contrast as long as their RMS contrast is equal. RMS contrast does not depend on spatial frequency contrast of the image or the spatial distribution of contrast in the image. All the faces are maintained with the same illumination and same RMS contrast where α is the contrast and β is the brightness to be increased or decreased from the original image f to the new image g as in Equation 2. On the other hand, histogram equalization is used to compensate the lighting conditions and enhance the contrast of the image. This is due to the face images may encounter poor contrast because of the limitations of the lighting conditions especially indoor.

21

2

1)(1⎥⎦

⎤⎢⎣

⎡∑ −==

−n

ii xx

nRMS (1)

βα += fg (2)

Eigenfaces Current real time recognition solutions employ techniques like Viola-Jone algorithms, Motion History Image (Huang and Lin 2008), Skin Color Model (Huang and Lin 2008), Support Vector Machine, SVM (Tistarelli et al. 2009), Neural Network or Active Appearance Model, AAM (Datcu and Rothkrantz 2007). There are many algorithms for real time face recognition and most of them depend on the features on the face to locate the global position (the face) and also the local position (features on the face). These algorithms are complex and unable to provide a universal solution and they require higher computational time. Some other solutions proposed a feature extraction module that estimates only the state of facial features. Then intelligent classifiers are designed to cater for the inherent uncertainty in the estimated facial features. However, these algorithms fail when some facial features cannot be detected or hidden by accessories. Furthermore, in a fully unconstrained environment, frontal faces may be taken under different illuminations, from varying distances and directions, in cluttered backgrounds, and with unpredictable facial expressions (Raquel 1995).

Hence we adopt the holistic or appearance-based method. The whole face region is used as a raw input for the recognition process. Many researches based on this principle are combined with several other techniques. One of these techniques is projecting face images or sub images onto a low dimensional feature space using Principle Component Analysis (PCA). The face on the image can be represented by a vector containing the gray levels of the pixels and after that PCA is applied to capture the face space in a low dimensional space spanned by orthogonal eigenvector. A linear combination Eigenfaces are computed from these eigenvectors and then used to represent the faces. The typical technique is Viola and Jones algorithm embodied with 3 small steps: Haar-like features, integral image and AdaBoost learning rule (Boumbarov et

34


al. 2007). Viola and Jones’s main idea is to combine weak classifiers based on simple binary features which can be computed extremely fast. Simple rectangular Haar-like features are extracted; face and non face classification is done using a cascade of successively more complex classifiers that discards non face regions and only sends face-like candidates to the next layer’s classifier. Haar like feature has achieved a better performance compared to other methods such as Gabor wavelets due to its much lower computation cost. Basically, Haar like feature method comprises 2 steps: thousands of Haar like features are extracted in each frame and the same Haar like features in the consecutive frames are combined as the dynamic features. This means that every dynamic feature units are feature vectors. These features then can be coded into binary patterns.

This coding scheme has 2 advantages: it is easier to construct weak learners for AdaBoost learning with a scalar feature than with a feature vector and the proposed binary coding is based on the statistical distribution, so the encoded features are robust to noise. Each layer’s classifier is trained by the AdaBoost learning algorithm. AdaBoost is a boosting learning algorithm that can fuse many weak classifiers into a single more powerful classifier (Yang et al. 2009). Most of the recent researches use only Viola and Jones algorithm with some configurations to detect the face from the image (Yang et al. 2009, Geetha et al. 2009) and then utilize AAM to extract the face out of the background. Viola and Jones algorithm has a big drawback despite many advantages like against illumination condition and scale invariant that is it cannot detect rotated frontal face (Huang and Lin 2008).

Figure 2 Viola and Jones Algorithm

In our case of disabled, we could not focus only on ideal conditions where the individual subject is aware of the face recognition process and willing to cooperate to some extent by following simple guidelines that improve the chances of gaining access to the desired facility. Our emotion recognition system depends on the level of detail in facial features solely. Though detailed facial features give good recognition, they require tedious image preprocessing and filtering which increase CPU time in return. When a person expresses an emotion such as happy, angry or surprise; each point on the face has a relationship in distance corresponding to the emotions. By calculating the distance of every point on the face to others and comparing these distances with the distances of a neutral face, we can determine which expression is shown on a face. As none of the methods has perfect performance in most cases; a hybrid real time emotion recognition system is proposed. The combination of Viola-Jones is used to detect frontal face from video which integrates Haar-like features, integral image and AdaBoost learning rule. Principal Component Analysis that uses Eigenface to detect emotion shown on the faces is applied. With the frontal face detection, Viola and Jones algorithm is used. Our real time emotion recognition system includes 3 main modules. First, detect face on real time video recording; then extract a face from the video and transform the face into necessary parameter to supply to the last step i.e. classification emotion.

Back propagated Neural Network (NN) We employed a multilayer perceptron back propagated neural network to train the expressions as in Figure 4. The input layer receives Eigen vectors detected as its input. The number of nodes in training layer equals to the dimension of the features incorporating the Eigen vectors. The number of nodes in the output layer

35


equals to the number of individual faces the network is required to recognize. The number of epochs for this experiment was 10,000 and the goal was 0.01. The back propagated neural networking training algorithm is shown in Figure 3. In the initialization stage, all the weights and threshold values of the network are set to random numbers within (–Fi, +Fi) where Fi represents the sum of neurons, i in the network. In the activation stage, the network is activated by applying the inputs x1(t), x2(t),…, xn(t) and the desired outputs y1(t), y2(t),…, yn(t). The actual outputs in the training and output layers are calculated. In the weight training stage, all weights are updated, and the errors associated with the output neurons are propagated backward. Iteration, t is increased by 1. If termination does not occur, then the back propagation iterates again.

Figure 3 Back propagation algorithm

Figure 4 Back Propagated Neural Network

4. PROTOTYPE IMPLEMENTATION

Our prototype, EmoCom has been implemented with a training and detection interface. The training interface allows each user of the prototype to create their profiles of emotions. Our approach works for arbitrary user-defined emotion classes. However, for testing and evaluation purposes, we use the universal basic emotions. A profile requires a set of images showing various emotions to be fed into the system for training purpose. Once a profile is trained, the prototype is ready to detect the user’s emotion in real time. The real time mode capture images in two ways that is automatic and manual. For automatic detection, a time interval can be set for each profile. By default, the allowed classification is at 30 frames per second. Manual detection is for the user to test the system before really put into use for a particular profile. We assume a full frontal view of the face, but take into account our human subject dependent variations in head pose and framing inherent in video-based interaction.

EmoCom is portable to detect the users’ emotions by utilizing the built in camera on a laptop. It can be used at anywhere anytime as the illumination or lighting is not an issue for our face detection algorithm. EmoCom features an easy-to-use Graphical User Interface which allows the user to detect emotions, train profiles, and select profiles easily. A profile is a collection of images for a user to show his expressions and needs. Figure 5 shows the main interface, users can start the facial expression recognition

Yes

No

Start Initialization Activation Weight training

IterationTerminate?Stop

36


by clicking Start EmoCom button or perform other tasks by selecting the tasks’ corresponding shortcut from the Tools tool strip button located at the top-left corner. These checkboxes allow the user to choose the mode of the detection. Manual detection requires a user to click on the ‘Detect Emo’ button to detect the emotion while Auto Detect mode detects the emotion automatically according to the time interval defined by a user in the Time Interval Text Boxes. Time Interval Text Boxes are only enabled when the user check the Auto Detect checkbox. Start EmoCom Button will start the webcam, retrieve frame from the webcam for processing. The processing of these frames includes facial tracking and facial expression recognition. Stop EmoCom button stops the capturing and processing of the frames from the webcam and free the resources used. Detect Emo button is enabled when the user check Manual Detect button. Clicking this button will trigger the facial expression recognition and output a sound for the result of the recognition. Figure 5 also shows the profile management allowing a user to configure the profile settings. ‘Checked On Startup’ allows a user to define which profile is to be selected on the startup of the program File Path textboxes show the path of the trained data. Sound Path shows the path of the Sound Clips that will be played in the recognition phase. Users can personalize the output sound of the recognition phase by putting their recorded sound clips in a folder under Sounds folder located in the program’s directory and type the path here. Reset to ‘Default Settings’ will reset the changes back to default settings but will not save the changes unless the user applies them. ‘Apply’ is disabled when the module starts; it saves the changes of the profile settings to a subject’s profile.

Figure 5 Main interface and Profile management module

Figure 6 shows Training Module from existing images allows the user to train new profiles using existing pictures. Supported image formats are BMP, DIB, JPEG, JPG, JEP, PNG, PBM, PGM, PPM, TIFF and TIF. Users can drag and drop files onto the list boxes of different emotions. Preview of the file entries will be shown in the picture box left of the list box. Remove Non-Image Entries button will check through all the entries in the list boxes and remove the entries which are not recognized or not supported. After dropping files onto the list boxes, the user can type in their profile name and filename to be used when saving the trained data. Process button will be enabled after user type in the Profile Name and Save As textboxes. After selecting, the entries in the list boxes will be filtered first (same with remove non-image entries button), process and save as .xml file. The Training Module from Camera is able to guide the users to train their profile using their own webcam. This training process involves 5 steps. Step 1 to 4 is taking pictures of different emotions for processing in Step 5. In each of these steps the user is required to take at least 5 pictures to ensure adequate recognition rate. Taking more pictures in these steps will increase the recognition rate. In Step 5 the user is requested to enter a profile name and the trained data will be saved using the profile name. After clicking the process button, the trained data will be saved as a .xml file. User is requested to enter profile details in the profile management window if the training is successful.

37


Figure 6 Training Module from existing images

5. PILOT TEST WITH TRAINING IMAGES FROM AT&T AND JFFE

The training data has been obtained from two face databases that are AT&T and JFFE before implemented on our targeted disabled community. AT&T has a total of 400 2D frontal face images from 40 human subjects. On the other hand JFFE has a collection of 213 2D frontal face images posed by 10 female human subjects. The AT&T database is tested first with our facial expression recognition using Eigen vector based back propagated Neural Network. AT&T database contains 10 different images of 40 distinct human subjects in 5 different illumination conditions. The image is resized to 256 x 256 pixels to maintain consistency to our second face database test. For some human subjects, the images were taken at different times, varying lighting, facial expressions and facial details (glasses/no-glasses). All the images are taken against a dark homogeneous background and the subjects are in up-right, 2D frontal position with a tolerance for some side movement. This mimics the context of our real environment setting where most of illuminations are due to sunlight and indoor lighting. The disabled are either in the care centre, home or school due to mobility restrictions. A 10% of the 400 images in the database were used as a training dataset and the remaining images were used as probe images in the facial expression recognition test. All images were convolved with Gabor filters. We used 10% that is 40 basis vectors of the 400 basis vectors representing 400 images. To each face image, the output equals to the number of individual faces the network is required to recognize which records the magnitudes of the Gabor filter response. The AT&T images are fully tested for face detection and facial expression recognition from database. The testing was performed with Pentium 4 3.00GHz CPU, 1 GB RAM, at the average running time for a face on an image to be detected and recognized (matched) is averagely 10 seconds for all the 40 human subjects tested. Table 1 shows accuracy of face recognition rate from each human subject. Averagely, AT&T achieves an average of 95.5% success rate.

The JFFE database is also used for pilot test our communication assistant as it contains more significant or stronger facial expressions. JFFE contains 213 images of 7 facial expressions posed by 10 Japanese female models. Each image has been rated on 6 emotion adjectives by 60 Japanese subjects. However we only selected 200 images by 10 models with 6 significant expressions each. We tag each expression for each human subject with a need. Hence the recognition of an expression triggers a message from the expression database to the respective caretaker. The average recognition rates from the JFFE, 96.5% is better than what we have obtained in the testing of AT&T database. The main reason is the facial expressions for each human subject are more distinctive causing less mismatched. However we do take into account the lack of gender difference in JFFE as it consists of all females. Furthermore, the database contains fewer images, 200 as compared to 400 2D frontal images.

Table 1 Pilot test results for 400 faces from AT&T and 200 faces from JFEE

AT&T JFEE Facial

expressions class

Average recognition rate

(%)

Standard Deviatio

n

Average recognition rate

(%)

Standard Deviatio

n

Happy 96.1 0.54 96.5 0.52 Sad 95.8 0.52 96.1 0.49

Scared 96.5 0.46 97.0 0.50

38


6. TESTING WITH REAL WORLD DATA

The experiment is conducted by using a laptop complete with built in camera, microphone and speaker. The processor is Intel Pentium Dual CPU 2 GHz with 1.43 GB of RAM. A set of 10 volunteers are invited to take the real time communication assistant evaluation with our prototype. Each subject has taken a set of 5 images for expressing one of the needs i.e. emotion. They displayed various ways of being angry, sad, happy and scaredy. The 20 images of each subject were used to create their own profiles under the category of emotion. Hence the communication assistant is trained with various emotions from different users. Figure 7 to Figure 10 show some of the training images of a volunteer on his emotions. The average detection time for a successful emotion is 13 ms using either automatic or manual detection mode. The real time recognition of one of the human subjects is shown in Figure 11. The highest recognition rate is 87% with a training set of 20 images for each emotion category. The scared and angry emotions tend to have higher detection rate due to the significant facial expressions shown on the human subjects.

Figure 7 Happy emotion trained in the profile

Figure 8 Sad emotion trained in the profile

Figure 9 Scared emotion trained in the profile

Angry 95.6 0.51 96.1 0.51 Mean 96.3 0.5 96.5 0.51

39


Figure 10 Angry emotion trained in the profile

Figure 11 Real time recognition of happy, sad, scared and angry emotion from a volunteer

The success rate of detecting various emotions depends very much on the training images for a particular expression for each human subject. We found out that 20 images for training an expression would give the highest recognition rate of 87% for all of our 10 samples. It is difficult and time consuming to obtain more images for profile training from the disabled for each emotion. They have problems focusing on the USB digital camera and show the same emotion for a required time. The detection using real time data has a low recognition rate due to insignificant expression for a particular emotion and insufficient training data being provided. Table 2 Communication Assistant recognition rate for emotion with 10 human subjects

Table 3 Comparison of confusion matrices of recognition for emotion with 10 human subjects

1-nearest neighbor classifier Angry .74 .04 .00 .07 Scared .04 .83 .00 .00 Happy .05 .00 .80 .03 Sad .08 .00 .00 .72 SVM classifier Angry .83 .02 .00 .04 Scared .04 .87 .00 .00 Happy .00 .00 .83 .03

Emotion Average detection % Standard Deviation Happy 85 0.51

Sad 83 0.53 Angry 85 0.49 Scared 87 0.48 Mean 85 0.5

40


Sad .08 .00 .00 .80 Haar + AdaBoost Angry .85 .03 .00 .03 Scared .05 .87 .00 .00 Happy .00 .00 .85 .01 Sad .09 .00 .00 .83 Angry Scared Happy Sad

7. CONCLUSION

For the initial stage, our communication assistant, EmoCom is developed to recognize the first aspect of needs expression that is emotions of the disabled who could not make sensible speech and physical movements. They require full time assistance in most of their physical movements. Hence we utilize the state of the art of information communication technologies to recognize their needs through expressions shown on the faces. At the following stage, we will work on recognizing the expressions on other needs such as food, toilet, like, dislike, games and movement. In recognizing other needs, we anticipate the problem of differentiating the expressions of various disabled users with different level of severity.

8. REFERENCES

AT&T FACE DATABASE (The ORL Face Database), http://www.cl.cam.ac.uk/Research/DTG/attarchive/pub/data/att_faces.zip. Accessed on 31 December 2008.

BOUMBAROV, O., SOKOLOV, S., GLUHCHEV, G.: Combined face recognition using wavelet packets and radial basis function neural network. CompSysTech '07: Proc of International conference on Computer systems and technologies. ACM (2007).

DATCU, D. AND ROTHKRANTZ, L.: Facial expression recognition in still pictures and videos using Active Appearance Model. A comparison approach. ACM (2007).

EKMAN, P.: Emotions Revealed. First Owl Books, New York: Henry Holt and Company LLC (2004). EKMAN, P.: Facial Expressions, in Handbook of Cognition and Emotion, T. Dalgleish and M. Powers,

Editors. John Wiley & Sons Ltd (1999). GEETHA, A., RAMALINGAM, V., PALANIVEL, S. AND PALANIAPPAN, B.: Facial expression

recognition – A real time approach. Expert Systems with Applications, Vol. 36(1): 303-308 (2009).

HEISELE, B., HO, P., WU, J., POGGIO, T.: Face recognition: component based versus global approaches. Computer Vision and Image Understanding., Vol. 91:1/2, pp. 6-21 (2003)

HUANG, X. AND LIN, Y.: A vision-based hybrid method for facial expression recognition. In the proceedings of the 1st international conference on ambient media and systems (2008).

JAPANESE FEMALE FACIAL EXPRESSION (JFFE) Database, http://www.kasrl.org/jaffe_download.html. Accessed on 31 December 2008.

RAQUEL, A.R.: Real-time Face Verification. Thesis (M.S.) Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science (1995).

TISTARELLI, M., BICEGO, M., GROSSO, E.: Dynamic face recognition: From human to machine vision. Image and Vision Computing, Vol. 27(3): 222-232 (2009).

WEYRAUCH, B. AND HUANG, J.: Component-based Face Recognition with 3D Morphable Models, in proceedings of 4th Conference on Audio- and Video-Based Biometric Person Authentication, pp. 27-34 (2003).

WONG, J.J. AND CHO, S.Y.: A local expert organization model with application to face emotion recognition. Expert Systems with Applications, Vol. 36(1): 804-819 (2009).

YANG, P., LIU, Q., AND METAXAS, D.N.: Boosting encoded dynamic features for facial expression recognition. Pattern Recognition Letters, Vol. 30 (2): 132-139 (2009).

ZANA, Y., ROBERTO, M. C. Jr.: Face recognition based on polar frequency features. ACM Transactions on Applied Perception (TAP), Vol. 3:1. ACM (2007).

Zhao, W., Chellappa, R., Phillips, P. J., Rosenfeld, A.: Face recognition: A literature survey. ACM Computing Surveys (CSUR), Vol. 35:4. ACM (2003).

41


Utilizing Semantic Web as Communication Protocol in Faded Information Field (FIF) Architecture for Information Retrieval and Dissemination

M. Ahsan Chishti * and Shaima Qureshi* Department of Computer Science & Engineering A. H. Mir Department of Electronics and Communication Engineering Shariq Haseeb Next Generation Networks Iftikhar Ahmad Defense and Systems Institute Abstract

Even though plenty of information is available on the Internet, still there is too little useful information there. For the Internet user one of the important parameter of the network is the speed of the retrieval of files. Since internet is rapidly evolving, new information gets added and modified continuously, thus there is the need to sort, organize and retrieve these resources so as to meet the user’s heterogeneous requirements fast and effortlessly. In this paper, a content code is proposed for faded information field (FIF), known as semantic web. This will act as a communication protocol for information transmission and a method of communication among the nodes in the network of FIF. Categories and Subject Descriptors: C. [Computer Systems Organization]: C.2 Computer-Communication Networks - C.2.1 Network Architecture and Design General Terms: faded information field, mobile agents, semantic web, information retrieval, autonomous decentralized system. _____________________________________________________________________________________

IJCIR Reference Format: M. Ahsan Chishti, Shaima Qureshi, A. H. Mir, Shariq Haseeb and I. Ahmad. Utilization of Semantic Web as Communication Protocol in Faded Information Field Architecture for Information Retrieval and Dissemination. International Journal of Computing and ICT Research, Vol. 3, No. 2, pp. 41- 46. http://www.ijcir.org/volume3-number2/article5.pdf. ______________________________________________________________________________________ 1. INTRODUCTION

In the past information was limited and the distribution of information was also very much restricted. With regards to that time, much was need for organization of information. But nowadays the information flow is

* Author’s Address: M. Ahsan Chishti and Shaima Qureshi, Department of Computer Science & Engineering; A. H. Mir, Department of Electronics and Communication Engineering; Shariq Haseeb, Next Generation Networks; Iftikhar Ahmad, Defence and Systems Institute.

"Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than IJCIR must be honored. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee." © International Journal of Computing and ICT Research 2009. International Journal of Computing and ICT Research, ISSN 1818-1139 (Print), ISSN 1996-1065 (Online), Vol.3, No.2 pp. 41-46, December 2009.

42


tremendous. The advancement in the field of network technologies has made it impossible to cover the overall information available. New sources of information are added continuously which are of many forms such as HTML, PDF, XML etc. To retrieve any information on internet, the process is not efficient as the information retrieved is not always the one which is required. Thus, there are lot of problems in information search, extraction, representation and interpretation.

Enhancement of information dissemination and retrieval will be of great help for researchers, developers, in addition to corporate, government, healthcare, and higher education. This effective information retrieval can be done through the use of such a network wherein it would be easier for a user to extract information with ease, fast and also efficiently. In this paper, a mechanism is proposed with the help of push and pull mobile agents. The user can consistently obtain the required information at the local nodes in the Faded Information Field (FIF) which uses the semantic web protocol of information broadcasting.

A new approach, know as semantic web, has been evolving which brings powerful AI concept in contact with the web infrastructure. Semantic web is envisioned as an extension of the current web where documents are annotated with meta-information [Davis et al. 2003]. It has the ability to define and link web data in a way that it can be understood and used by machines for automation integration and reuse of data across various applications. Semantic web is the presentation of machine-processable semantics of data on the web. It is a collaborative effort led by W3C Consortium with participation of large number of researchers and industrial partners [Berners et. al. 2001; Jeckle and Zhang 2003].

In essence, the solution to the above problem is based on an approach which links the semantic web with faded information field. This solution is applied on a decentralized network, thereby facilitating in standardizing the information distribution across the network. A lot has been written about the use of FIF [Ahmad et. al. 1999; Arafaoui and Mori 2000; Arafaoui and Mori 2001] but no communication format or the method by which nodes of a FIF network can communicate has been forwarded. This paper describes the use of semantic web technology for use in faded information field architecture in order to create and deliver the technology with Information provision and utilization.

2. FADED INFORMATION FIELD ARCHITECTURE (FIF)

A Faded Information Field is a set of computers around an information provider where the information provider distributes its information contents. Thus, the information of the provider is replicated to surrounding nodes, creating a “field” of information.

The faded information field can be constructed to balance the cost of information allocation performed by push-Mobile Agents and the cost of information retrieval performed by pull-Mobile Agents. Fading improves the reliability of the information in the field, and effectively reduces the burden on the server and the network of maintaining the information field. Conversely, it will mean that some agents will have to travel further in the network to discover the required information [Arfaoui and Mori 2001]. The main goal of the faded information field is to guarantee the assurance of autonomous service provision and autonomous service utilization [Ahmad et. al. 1999; Mori 1999]. The main goal of FIF is to guarantee the assurance of autonomous service provision and autonomous utilization [Arfaoui and Mori 2001]. The trend for information distribution is given in Figure 1. The graph indicates the decrease in information as the distance from the main server. Thus, the nodes which are near-by to the server have more information and in turn are updated more frequently than the nodes which are at a greater distance.

Figure 1: Information distribution in FIF

Am

ount

of f

orm

atio

n

Distance from service provider

43


2.1 Push Mobile Agents and Pull Mobile Agents (PMA’s) A mobile agent is a software program that can be transferred from one location to another in a network environment. The agent is self-sufficient to make decisions with reference to its execution at a host, and it may choose to save its state, move on to another host, and then continue execution. A mobile agent does not need to maintain communication with its source. Therefore, it is an autonomous entity.

Figure 2: Network Structure and Mobile Agents

By the use of mobile agents, access time for the user to get the appropriate information is improved because the user need not reach the service provider and it can get its required information at the nodes which are close to its premises as shown in the figure 2, there by enabling Service Provider (SP) to avoid the congestion. The Mobile agents can support each of four main monitoring activities performed in an object based system – generation, processing, dissemination and presentation of information [Ahmad et. al. 1999; Sloman 1994]. Mobile agents have a complex nature with respect to their design, implementation and maintenance [Liotta 1999]. 2.2 Communication format for FIF The data is usually sent by conventional methods where in the destination methods are mentioned in the packet. But in FIF, communication is done using the content codes. Content codes (CC) are uniquely defined with respect to the content of the information service. This information is further divided into characteristic codes (CH). CH defines the properties corresponding to CC. Pull mobile agents requests in terms of this CC and CH, thereby enabling pull MA to broadcast [Arfaoui and Mori 2000].

As an example, consider that a user needs information regarding buying a house. The CC would be the purchase of house, and the corresponding CH’s can be the place where the user wants to purchase the house, the price, type of house etc. Depending upon the attributes of the search, provided by the user, this format (as depicted in Figure 2) will try to look for any advertisement for sale of house on the web. This streamlined search provides the ultimate and most useful information requisite for the user. House Place KL Price 200000 Type Link

CC1 CH1 Data CH2 Data CH3 Data Figure 3: Message Format in FIF

Thus, owing to the fact that the query in faded information field would have the CC and CH’s, so the full knowledge format requires certain specific protocol by which this communication can be done. The Semantic Web will bring structure to the meaningful content of Web pages, creating an environment where software agents such as those above mentioned mobile agents roaming from page to page can readily carry out sophisticated tasks for users. As for example the search that we have mentioned above is looking for the information and in Semantic Web format, the whole detail will be provided starting from the location of the house, its price, type and in addition to this certain keywords such as "balcony, swimming pool, rooms”, which can be mentioned in the CH’s will help in retrieval of specific and more comprehensive data [Arfaoui and Mori 2001].

44


3. SEMANTIC WEB

To reuse data across the internet, a common framework is needed which allows data to be shared and reused. Ontologies play a major role in supporting information exchange across various networks. Currently, Ontologies applied to the World Wide Web are creating the Semantic Web. Semantic Web is a collaborative effort led by World Wide Web Consortium (W3C) with participation from a large number of researchers and industrial partners. W3C develops interoperable technologies (specifications, guidelines, software, and tools) to lead the Web to its full potential. It is a forum for information, commerce, communication, and collective understanding. Semantic Web is based on the Resource Description Framework (RDF), which integrates a variety of applications using XML (Extensible Markup Language) for syntax and Universal Resource Identifiers (URIs) for naming [W3C 2004; Fensel 2001]. As shown in Figure 3, RDF builds on XML to create descriptions, and descriptions are metadata, that is data about data. It allows anyone to design their own document format and then write a document in that format. RDF is a format to make statements that are meant to show something. It is identified by a unique address called the URI.

According to Tim Berners-Lee, James Hendler, Ora Lassila "The Semantic Web is an extension of the current web in which information is given well-defined meaning, better enabling computers and people to work in cooperation" [Berners-Lee et. al. 2001]. The Web can reach its full potential only if it becomes a place where data can be shared and processed by automated tools as well as by people. For the Web to scale, tomorrow's programs must be able to share and process data even when these programs have been designed totally independently. The Semantic Web is a vision: the idea of having data on the web defined and linked in a way that it can be used by machines not just for display purposes, but for automation, integration and reuse of data across various applications [W3C 2004].

4. INFORMATION RETRIEVAL AND FILTERING TECHNOLOGY FOR FIF

RDF is a language intended to be used to express propositions using precise formal vocabularies for access and use over the World Wide Web. Technically speaking, Web-enabled languages and technologies are being developed (e.g. RDF-Schema, DAML+OIL, DAML-Rules, Rule-ML), schema and ontology integration techniques are being examined and refined and Web Services Integration Standards are being defined (e.g. UDDI, JINI).

Instead of developing own ontology from scratch, it is suggested to find an existing ontology that is broadly accepted [Abramowicz 2003]. RDF combines terms into triples, sets of three which express basic concepts or statements.

Figure 3: The Semantic Web layers The Semantic Web adds a type to this link, making it expressed more efficiently, such as the case of retrieval of information regarding house as mentioned in section 2. The query would be expressed as (House, Price, Location). In RDF it can be expressed as:

Universal Resource Identifiers (URIs)

HTML +XML

RDF

Ontologies

45


<house> <name> "house" . <kl> <location> <kl> . <200000> <price> <200000> . <Web> <website> <http://www.mzchishti.com/> Semantic Web combines URIs, HTTP and RDF to build a system of machine-to-machine communications for sharing information. Take the example of purchasing a house. To begin with, we need the information about the house; <House> <...location> "KL" . <House> <...price> "200000" . <House> <...type> "link" . ... Thus, we can ask for the query for the house as: < http://www.mzchishti.com /> <...wantshouselocatedat> <kl> . < http://www.mzchishti.com /> <...withprice> "200000" . < http://www.mzchishti.com /> <...housetype> "link" . ... By using RDF name space, the information can be shown in simpler way: @prefix : <http:// www.mzchishti.com /rdf/> . @prefix h: < http://www.mzchishti.com /info/> . h:1000 :type :Info . h:1000 :enquiry "house" . h:1000 :price "200000" . h:1000 :type "link" . … Ontologies are key enabling technology for the Semantic Web. In making Semantic Web as capable communication protocol in FIF architecture requires acquiring ontologies and linking them with data. An RDF schema acts as a repository that provides the storing and maintaining ontologies and their instances. The last step would be querying and browsing semantically enriched information sources. This technology can also be used for electronic knowledge sharing and reuse. This offers a heterogeneous representation of web resources, thereby enabling an extremely knowledgeable system with specialized services [Abramowicz 2003]. Thus, FIF would act as a semantically enriched search engine, browsing and providing support for sharing.

5. EXPLOITING SEMANTIC WEB FOR FIF: A FUTURE PERSPECTIVE

Faded Information Field architecture brings flexibility, reliability and real-time property to the information environment. This research mainly gives a method and is based on a prototype of FIF but the methodology needs to be established well. Effective and efficient protocol for communication in FIF requires an advanced ontology for expression and representation.

According to TimBL, there will be many layers to the Semantic Web, which could take around ten years to complete:

• Unicode and XML • RDF and other Basic Assertion Languages • Schema Langauges • Conversion Language

46


• The Logical Layer • A Proof Language • An Evolution Rules Language • Query Languages for Proof Validation (Swartz, 2006).

A well defined formal semantics has to be established to ensure interoperability. The effectiveness of this protocol will enable the mobile agents, which have been designed specifically for FIF, to work together thereby helping in transferring of data among different MAs. Much of the current work on the Semantic Web centers on a variety of technologies that are already in widespread have to be practical in use, particularly the Resource Description Framework (RDF)—which lets content creators express structured metadata statements describing URIs.

6. CONCLUSION

Faded Information Field has been designed to suit the requirements of service providers and the users as well. Semantic web technology can make an important contribution in acting as communication protocol. Semantic web can work across different networks, thereby helping in communicating across the wide range of networks.

This paper clarifies the rationale that would enable effective, efficient and robust communication between mobile agents in FIF. We have explained the concept and realization of effectual methodology for FIF communication with the help of semantic web for Push/Pull mobile agents.

7. REFERENCES

ABRAMOWICZ W., Knowledge-Based Information Retrieval and Filtering from the Web,” Kluwer Academic Publishers, USA, 2003.

AHMAD H., ARFAOUI H. AND MORI K., “Autonomous information fading by mobile agents for improving user’s access time and fault-tolerance,” Proceedings FTDCS, pp. 297-283. IEEE, 1999.

ARFAOUI H. & MORI K., “Autonomous Navigation in Information Systems for Load Balancing User Demands,” Proceedings of the Fifth International Symposium on Autonomous Decentralized Systems, 2001.

ARFAOUI H. AND MORI K., “Autonomous Integration of Information Services In Heterogeneous FIF System,” Proceeding of International Workshop on Autonomous Decentralized System (IWADS 2000), pp.40-45, Sep.21-23, 2000.

BERNERS-LEE T., HENDLER J., LASSILA O., “The Semantic Web,” Scientific American, May 2001. DAVIS J., FENSEL D. AND HAEMELEN F. V., Towards the Semantic Web, John wiley & Sons Ltd,

England, 2003. FENSEL D., HARMELEN F. V., HORROCKS I., MCGUINNESS D. L., PATEL-SCHNEIDER P. F.,

“OIL: An Ontology Infrastructure for the Semantic Web,” IEEE Intelligent Systems, pp. 38-45, Volume 16, 2001.

JECKLE M. AND ZHANG L. J., “Semantic Web Enabled Web Services: State-of-Art and Industrial Challenges,” ICWS-Europe 2003, LNCS 2385, pp. 183-197, 2003.

LIOTTA A., KNIGHT G. AND PAVLOU G.. “On the performance and scalability of decentralized monitoring using mobile agents,” Proceedings of 10th IFIP/IEEE International Workshop on Distributed Systems, 1999.

MORI K.. “Autonomous fading and navigation for information allocation and search under evolving system,” Proceedings APSITT, pp. 326-330, IEEE, 1999.

SLOMAN S.. Network and distributed Systems management. Addison-Wesley, pp 303-347, 1994. SWARTZ A., “What is the Semantic Web?,” SWAG - Semantic Web Agreement Group. [online]

http://swag.webns.net/whatIsSW. Retrieved on June 2006. WORLD WIDE WEB CONSORTIUM (W3C), “Semantic Web,” [online] http://www.w3.org/2001/sw.

2004.

47


An Information Databank Framework for the Health Care Industry in Nigeria

OLUTOLA .M. OBEMBE* LM Ericsson Nigeria OLORUNTOBA .S. OGUNDELE Department of Computer Science, Federal University of Technology, Akure, Nigeria Abstract

The development of web based database systems for service industries has led to a remarkable new dimension of information retrieval and distribution. These developments allow service industries to make information readily available almost at no cost to the general public. This paper discusses the use of web based database systems to accommodate the health care industry data. Presently, the health care information system in Nigeria consist of series of scattered inter-related data from different health sources and service centres, The Internet being a global village was used as a tool to solve this information problem by capturing all these related data into a data repository online where it can be accessed as information. The design uses the three-tier web model architecture as its underlying technology and presents an Information databank capable of storing health care data. This system when deployed will be advantageous in that it saves time and resources for both health administrators and the people; it also helps with health statistics and opens a new market trend for the health industry. However, since this is a prototype design, major improvements such as increased infrastructure, query optimization need be done to accommodate large scale deployment.

Categories and Subject Descriptions: H 3.5 [Information Storage and Retrieval] – Online Information Services – Web Based Services. General Terms: Design, Database System. Additional Key words and Phrases: Databank, Health, Health Care System, Nigeria

IJCIR Reference Format: OlutolaM. Obembe and Oluruntoba S. Ogundele. An Information Databank Framework or the health Care Industry in Nigeria. International Journal of Computing and ICT Research, Vol. 3, No. 2, pp. 47-57. http://www.ijcir.org/volume3-number2/article6.pdf. 1. INTRODUCTION

The Health Care Industry in Nigeria unlike developed countries in Europe and North America is a growing one. This is obvious in the unavailability of medical care, functional health care system and the attention

* Author’s Address: Olutola M. Obembe, LM Ericsson Nigeria, 17 Walter Carrington Crescent, Victoria Island, Lagos, Nigeria. [email protected]; Oluruntoba S. Ogundele., Department of Computer Science, Federal University of Technology, Akure, Nigeria., [email protected]

"Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than IJCIR must be honored. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee." © International Journal of Computing and ICT Research 2009. International Journal of Computing and ICT Research, ISSN 1818-1139 (Print), ISSN 1996-1065 (Online), Vol.3, No.2 pp. 47- 57, December 2009.

48


paid on health services by the public and private sectors. However, in the last decade, major improvements were witnessed in this sector in Nigeria and Africa at large. These improvements are made evident in the restructured modalities of the Nation’s Health Care systems vis the changing pace of health services delivery, increased competence of Medical Practitioners, availability of medical resources and many others.

Health care information system has been defined by Shortliffe and Perreault (2001) as a field of study concerned with the broad range of issues in the management and use of biomedical information, including medical computing and the study of the nature of medical information itself. If physiology literally means ‘the logic of life’, and pathology is ‘the logic of disease’, then health informatics is the logic of healthcare which is the science that studies the use and processing of data, information, and knowledge applied to medicine, health care and public health (van Bemmel and Musen, 1997).

The health care system forms a major building block of the Health care industry and some of the goals of the health care systems discussed in (WHO, 2000) are good health, responsiveness to the expectations of the population, and fair financial contribution. Duckett (2004) proposed a two dimensional approach to evaluation of health care systems: quality, efficiency and acceptability on one dimension and equity on another. The basics of Heath Care System (HCS), however, remain the same - national coverage for medically necessary health care services provided on the basis of need and quality, rather than the ability to pay.

Two major models were identified according to Wikipedia Encyclopedia, (2008) that have been adopted in the design of health care systems: private and public models. Private enterprise health care systems are comparatively rare. Where they exist, it is usually for a comparatively well-off subpopulation in a developing country with a poorer standard of health care. For instance, private clinics established primarily to cater for a small, wealthy expatriate population in an otherwise poor country. The other major models are public insurance systems. Instances of these could be social security, publicly funded health care, social health insurance health care model and so on; where workers and their families and other country residents are insured by the government.

The basic functional areas of a successful health care industry are e-Health, Funding, Health Care Delivery System, Home & Continuing Care , Hospital Care , Nursing Policy, Palliative & End-of-Life Care , Pharmaceuticals , Primary Health Care , Legislation & Guidelines , Reports & Publication and Health Services just to mention a few. While the strength of these areas is not evident in the present Nigeria health care system, it is worthy to note that the present government is trying to improve the health system of the nation.

This paper’s concern is not on the basis or position of the current health care system as it were, it focuses on the issues affecting the availability of health resources, medical expertise, and medical services which includes lack of adequate information, lack of awareness, and proper orientation in the health industry within Nigeria. Thus, the proposed objective is identifying the causes of this problem and proffering a solution through the design of a computer based information system. A database is a collection of interrelated data while a databank can be considered as a storehouse containing related data, consequently both terms will be used interchangeably within the context of this design. 1.1 Problem Definition

In Nigeria, several factors undermine the effective delivery of the Health Care System. These include lack of funds, lack of health services, insufficient expertise/professionals and above all lack of information of the few places where particular professional services are being delivered. On the issue of Inadequate Information which is what this paper addresses, Patients and people needing health services are sometimes driven into uncertainty when they are in need of some particular services. It is an accepted fact that the nations health system does not have everything yet but the problem is not really the inadequacy of the health services available, it is more of lack of information of where they can access the little services available. However, it is not sufficient to say that the nations system does not have this data; it is more of a problem of how majority of the people can have access to this information. The media, advert houses and the newspapers have played their role but as there is limit to their capacity in providing the required health information to the citizen. Thus, how do we decentralize this few health service information and made it available to the general public?

49


2. A HEALTH CARE SYSTEM INFORMATION DATABANK

The use of the Internet to solve the growing information gap between the limited resources available and the people cannot but be over-emphasized. This System will allow the development of a database system where available health services (both old and new) are made known to the public via a web front. The users of this system being the people, patients, health professionals, and government’s e.t.c can make quick enquiries as to the availability of health care systems, health manpower (Doctors, Pharmacists, and Specialists), hospitals, Drug Stores and so on nearest to them. This in turn will break the barrier of lack of information to the people and also help the federal government to make available services to the people while the health Industry catches up with its expected goal of providing correct and good health services in every neighborhood.

Fig 1: Schematic Representation of the Proposed System. Figure 1 shows a brief relationship between the Unified Data Repository which contains the details of the health services available, health human resources, locations, health schemes, the health data and its users, primarily the people (the public) who need this information. The question arises of how the Databank gets its data? This is provided by the health service providers, health practitioners, the Administrative body in charge of this Information System and a web service technology discussed in later sections of this paper.

3. WEB DATABASE SYSTEMS

The development of dynamic web pages or Web Applications over the Internet has been in operation for over two decades now, This was engineered by the introduction of web scripting languages like cold fusion, PHP, ASP which are used to write application programs that run on the internet and the possibility of developing three-tier applications using databases on the Internet.

Fig 2: Three (3)-Tier Web Model

DataSource Information

Request

Client 1

Client 2

Client N

Database System

Client

Web Server

DB Systems

Client 1

Client 2

Client N

Browser 1

Browser 2

Browser N

Web Server

<HTML>

HTTP Query

Data

Tier 1 Tier 2Tier 3

50


Database systems are designed to manage large bodies of information. According to Silberschatz-Korth-Sudarshan (2004), management of data involves both defining structures for storage of information and providing mechanisms for the manipulation of information. In addition, the database system must ensure the safety of the information stored, despite system crashes or attempts at unauthorized access. If data are to be shared among several users, the system must avoid possible anomalous results. Tarricone and Esposito (2004) described the 3-tier architecture above as having the following;

• The front-end tier, the client from which the end user expresses requests by using a browser; • The middle tier, the Web server equipped with software capable of building HTML pages on the

fly, starting from data originated by users as well as data fetched from the backend tier; • The backend tier, a backend database where data are stored. The development of HCS databank will use the Web Model discussed above. While, the front-end tier

is concerned by the users and can be used as long as a user is connected to the internet, Attention will be on the Tier2 (Web Server, Web Pages, Application Languages) and the backend tier where the database model and structure of the application will be designed. For the purpose of this application, the Server-side technology used is PHP (PHP Hypertext Preprocessor). It is used in conjunction with HTML to render the web pages on the web. The database system used in MySQL, this is used because of its availability, cost and operational speed.

4. ANALYSIS OF THE PRESENT SYSTEM

The present health care system in Nigeria consists of both public (government) and private institutions. The Ministry of Health is in charge of the public health services provided by the government. Recently health care managers were introduced in Nigeria health schemes. These include National Health Insurance Scheme (NHIS), HYGEA and many others. In the private Sector, Health Institution, Clinics, Blood banks, Pharmaceuticals companies, Drug Stores, Laboratory, Specialist and Doctors all contribute to the success of health services within the country. There exist professional bodies such as Nigeria Medical Association (NMA) for Doctors and Specialist, National Association of Resident Doctors of Nigeria (NARD) (www.nigerianma.org), Nigeria Pharmaceutical Association (NPA) for pharmacists all registered with the aim of fostering common cause as it may affect the health care industry in Nigeria. There is no medium or facility on ground for the acquisition and repositioning of these inter-related data. 4.1 Proposed System Design and Architecture This system involves the design of a databank that unifies the scattered data of health inter-related services, resources, expertise, funding and so on. In particular, it is designed to house data related to Hospitals, Pharmaceuticals, Laboratories, Blood banks, Special Clinics, Doctors, Health schemes and the services rendered/offered by this categories. The system is designed as a web based system and is proposed to be administered by the Ministry of Health because of her major role in the Nations Health Care System.

The system design uses both the human-centric and application-centric web approach as described in (Codd, 1970). This allows for data to be inputted into the system by the administrative body and also for the application to source for data itself using the XML- service, This is to increase data authentication and integrity, this technology will be employed to retrieve correct and up-to-date data from National Associations e.g. Nigeria Association of Doctors for accredited Doctors and their areas of specialty.

The system architecture below shows the blueprint design of the Information Databank and the flow of data storage and retrieval within concerned entities. In the architecture below, sub section A & B are human centric in that they require input by humans to the web server while sub-section C is application-centric, this involves the web server communicating with another web server for information, both sending and receiving data and information via XML web service.

Section A of the architecture involves two kinds of users. The first sets of users of the system are those who place information request to the system. Their request is sent via the HTTP query to the web server, which in turn fetches the right information and displays the needed information back to their screen. The second sets of users are health care professionals, health service centers e.t.c who register their services and contact within the system. Section B involves the Administrative users, while they can also be seen as query users. Their functions within the system include authenticating registered members and inputting of data into the database via the web server. Their request is also sent via the HTTP query to the web server which in turn performs the query operation and delivers the information requested. Section C involves the

51


Information System web server communicating with other database system via an XML-web service. The Web server sends XML request as described in Cerami (2002) to the outside web server (this outside web servers being web servers of Health Services body like NMA, NARD, NPA where authentic information about health professionals can be gotten) and receives an XML-response with data which is then used to update its own database. 4.2 Database Relational Design

Fig 3: Information Databank System Architecture

Section A of the architecture involves two kinds of users. The first sets of users of the system are those who place information request to the system. Their request is sent via the HTTP query to the web server, which in turn fetches the right information and displays the needed information back to their screen. The second sets of users are health care professionals, health service centers e.t.c who register their services and contact within the system. Section B involves the Administrative users, while they can also be seen as query users. Their functions within the system include authenticating registered members and inputting of data into the database via the web server. Their request is also sent via the HTTP query to the web server which in turn performs the query operation and delivers the information requested. Section C involves the Information System web server communicating with other database system via an XML-web service. The Web server sends XML request as described in Cerami (2002) to the outside web server (this outside web servers being web servers of Health Services body like NMA, NARD, NPA where authentic information about health professionals can be gotten) and receives an XML-response with data which is then used to update its own database. 4.2 Database Relational Design

HTTP Post/Get

HTTP

Query

HTTP Query Data

XM

L R

espo

nse

XM

L R

espo

nse

Client

Client

Client

Admin Client

Admin Client

Web Server Web Server

Web Server

Database

A

B

52


A relational database consists of a collection of tables, each of which has a unique name. Each table contains columns known as attributes and rows known as tuples which represent relationships among a set of values. According to (Codd, 1970), a Relational Database Model can be represented as: R0 = {A1, A2, A3, …., An-1, An } Where R represents a relations and A1, A2 represents the attributes contained in the relation R. Thus, using the above representation, the database design of the HCS Database with the relations is shown below. Hospital {hos_id, hos_name, hos_contact, hos_desg, hos_add, hos_country, hos_state, hos_zip, hos_url, hos_tel, hos_email} Pharmacy {pha_id, pha_name, pha_contact, pha_desg, pha_add, pha_country, pha_state, pha_zip, pha_url, pha_tel, pha_email} Doctor {doc_id, doc_fname, doc_lname, doc_ini, doc_degree, doc_practise, doc_practise_add, doc_city, doc_state, doc_zip, doc_practise_tel, doc_email, doc_url, doc_acc_patient, doc_pri_spe, doc_sec_spe, doc_ter_spe} Laboratory {lab_id, lab_name, lab_contact, lab_desg, lab_add, lab_country, lab_state, lab_zip, lab_url, lab_tel, lab_email} BloodBank {bb_id, bb_name, bb_contact, bb_desg, bb_add, bb_country, bb_state, bb_zip, bb_url, bb_tel, bb_email} Special-Clinic {clinic_id, clinic_name, clinic_contact, clinic_cont_desg, clinic_add, clinic_country, clinic_state, clinic_zip, clinic_url, clinic_tel, clinic_email} Medical_Service {medical_id, hos_id, med_echo_lab, med_eeg_lab, med_endoscopy, med_occ_health, med_xray. Med_radio, med_bb, med_unit, med_oxy, med_amb } Bb_blood { bb_blooddetail_id, bb_id, bb_blood_type, bb_avail } Capacity { capacity_id, hospital_id, cap_bed_type, cap_no_of_bed, cap_staff_str, cap_staff_str_sup} Drug_portfolio { drug_id, pharmacy_id, drug_name, drug_desc, drug_ind} Healthcare { health_care_id, hospital_id, hc_hygea, hc_nhis, hc_others} Lab_test { test_id, lab_id, test_name, test_desc} Specialist { specialist_id, hospital_id, spe_specialist, spe_paed, spe_gyna, spe_dent, spe_derm, spe_surg, spe_other } Surgical { surgical_id, hospital_id, surgical_proc} The database table relationship(s) is shown in appendix A. 4.3 Process Design of the System The process design in Fig 4 below shows a quick view of the application usage from the users’ point of view. The Index page which doubles as search query page is the application start phase. The Process flow shows two basic modules of the system, Module A for class A users who use the system basically to request information about health service and practitioners while Module B which is for Class B users (This class of users are the medical practitioners, health service providers who register to make their presence known within the system) is divided into two, New Users and Existing Users, New Users go through the registration procedures while Old users simply login into their member area to edit/update their information. The user information update is required from time to time to maintain an accurate data within the database system.

53


Fig 4: Diagram showing the process flow

As shown in the process flow diagram in figure 4, when a query for information is sent to the database system, the query is processed and subsequently returns a result page if query execution is true or a search error page if query execution is false. Existing Members gain access to their member area using their username and password, once logged in; they can edit and update their information within the database system.

This design was tested on a Pentium III, 900 GHz, 400MB Ram Computer machine with Microsoft Windows XP Professional as its Operating system and the simulation was demonstrated on a stand-alone system.

5. DEPLOYMENT STRATEGY AND APPLICATION USAGE

Just like any web application, this web system will be deployed over the Internet and can be hosted by any Internet web hosting service provider e.g yahoo, hostmonster, resellerguru e.t.c. Such Internet providers must have the capability of hosting PHP/MySQL Applications. Also, the database space must be taken into consideration as the system needs a relatively large database space due to the amount of data expected to be used and analyzed by this system. Once the system as been uploaded, it becomes readily available for use by the general public via the domain name registered for it..

The usage of the system is relatively easy; the index page shown below is the search page of the web system. It shows the search categories that can be used for query search, the register and login links for new and old users respectively and the search system itself.

Logi

n fa

lse

Login true Member Area Edit/Update Member Details and Services

New Member Registration

Login

Members

Index Page (Search Page)

Register Index

Search Result Page

Full Profile View

Login Error

Error Page (Null Result

Query False

Query True

Module A

Module B

54


Fig 5: Search Category Link

Fig 6: Member Links

Fig 7: Search Page (Index Page) Once a search category is highlighted and a search request is executed, the query analyzer analyzes the query and returns a result page if query was successfully executed with data like figure 8 below.

Fig 8: Search Result Page

55


Fig 9: Search Error Result Page The registration page is shown below; any new member registers under the appropriate category as it fits his/her contribution to the health industry. Once a successful registration has been done, a member area is shown to the user to modify and update details of services; this update of course is required from time to time to achieve an overall up-to-date system.

Fig 10: Registration links

Fig 11: Member Area 5.1 Security and Data Authentication The security of this web system is crucial because of its sector importance (health), while the security of the system online can be guaranteed relatively to a large extent with the use of a reliable dedicated server online to restrict access and database replication in case of database failovers. The issue of data security and authentication is a major one because this is the core of the information being passed to the public and

56


the higher the validity of the data, the more useful and reliable it becomes to the people without the goal of the project being jeopardized. In order to ensure data authenticity, registered users data are not allowed to be processed along search queries unless the data has been validated and verified by the Administrative Body from the proposed registration body of both health practitioners like doctors, pharmacist and health service providers like laboratories, blood banks. The web service structure of the application ensures that data replication is not allowed and also validates data by retrieving data from only specified and authorized data sources.

6. CONCLUSION

This paper presents a technology of improving the health care industry information system with the development of an information databank. Prototype databases was proposed and developed to house health system related data such as health practitioners, health service providers, health schemes e.t.c.; and make it available to the general public via an online system (the internet). If the system is well deployed and managed properly, it will present the capacity of the health industry to the people in a more clearer and readable format. The advantages of this application when implemented are enormous. These include the presentation of a system that can help the health industry save resources, help the people (public) save time and resources with the availability of timely information of what, where and how about health care services around them. It also presents the government and the health Industry administration a easier medium of gathering heath statistics both at the practitioner level and the service level. This system as it were also presents to the health industry market an avenue to render its services on a much larger scale and thus improving health businesses while bridging the gap between patients and health professionals. 6.1 Limitations of the System The system has been developed to be workable within the context of our present environment, however, there may exist some limitations in implementing some aspects of the system. The application-centric side of the project to retrieve data by itself will suffer setbacks until there exist a web server and web service infrastructure for its intended data sources online. Another setback may come with the issue of funding the project, As the project will turn out to be a national project, it will require huge investments to keep the system alive and working. 6.2 Future Research This project despite its relevance and present technology can be improved upon. Further development to improve the design include the need to house more health service data, the need to incorporate the usage of query optimization techniques to fast track query results, expanding of the project to form a specialize database for health services within the country. Another issue is the need to replicate and better optimize MySQL servers which houses the database since it will continue to grow in size so as to improve performance.

7. REFERENCES

CODD E.F., 1970, A Relational Model of Data for Large Shared Data Banks, Communications of the ACM, Vol. 13, No. 6, June 1970.

CERAMI ETHAN, 2002. Web Service Essentials, O’Reilly Publishers. DUCKETT, S. J. 2004. The Australian health care system. 2nd edition ed. Melbourne: Oxford University

Press. LUCIANO TARRICONE, ALESSANDRA ESPOSITO, 2004, Grid Computing for Electromagentics,

Artech House, Boston London. MARIO PIATTINI, OSCAR DÍAZ, 2000. Advanced Database Technology and Design, Artech House,

Boston London. NIGERIA MEDICAL ASSOCIATION website: http://www.nigerianma.org ROBIN MEANS, SALLY RICHARDS AND RANDALL SMITH (2008) Community Care: Policy and

Practice, Palgrave MacMillan. SHORTLIFFE EH, PERREAULT LE, 2001. eds. Medical Informatics: Computer Applications in Health

Care and Biomedicine. New York: Springer, 2001.

57


SILBERSCHATZ−KORTH−SUDARSHAN, 2004. Database System Concepts, Fourth Edition VAN BEMMEL JH, MUSEN MA, 1997. eds. Handbook of Medical Informatics. AW Houten, Netherlands: Bohn Stafleu Van Loghum; Heidelberg, Germany: Springer Verlag, 1997. WIKIPEDIA ENCYCLOPEDIA, 2008. Health Care System, retrieved May 6, 2008 from

"http://en.wikipedia.org/wiki/Health_care_systems“. WORLD HEALTH REPORT 2000 - Health systems: improving performance (WHO, 2000). 9. Appendix

10. Appendix a: examples of table showing relationship between attribute keys.

58


Effective Algorithm Using Similarity Based Technique for Browsing Image and Video Databases

Rekha B Venkatapur* A P S College of Engineering, Bangalore-78, Karnataka, India Dr. V.D.Mytri GND College of Engineering, Bidar, Karnataka, India Dr. A.Damodaram J N T U College of Engineering, Hyderabad, India Abstract

Navigator, an intelligent browsing-tool for image and video data is developed. The main problem with the handling of multimedia databases is the navigation through and the search within the content of a database. The major problem arises from the difference between the possible textual description (annotation) of the database content and its visual appearance. Overcoming the so called semantic gap has been in the focus of research for some time. In the present study, a new system for similarity-based browsing of multimedia databases is presented. The system aims at decreasing the semantic gap by using a tree structure, built up on balanced hierarchical clustering. Using this approach, operators are provided with an intuitive and easy-to-use browsing tool. In the present study an attempt is made is not only on the description of the database organization and retrieval structure, but also how the proposed techniques might be integrated into a single system. In this paper an attempt is made for the direct use of a balanced tree structure for navigating through the database of keyframes, paired with an easy-to-use interface, offering a coarse to fine similarity-based view of the grouped database content.

Key words: browsing, video databases, image databases, multimedia retrieval

IJCIR Reference Format: Rekha B Venkatapur, V.D.Mytri and A.Damodara. An Effective Algorithm Using Similarity Based Technique for Browsing Image and Databases. International Journal of Computing and ICT Research, Vol. 3, No. 2, pp. 58-64. http://www.ijcir.org/volume3-number2/article7.pdf. 1. INTRODUCTION

Sometimes a similarity search for a fix amount of "best matches" is not sufficient to get an overview of a video database. Especially when a user might not be satisfied with his results or when he wants to get a coarser idea of what could be inside a video collection. Nowadays, content-based image and video retrieval (CBIR/CBVR) are getting more and more into focus with the rapidly growing amount of image and video

* Author’s Address: Rekha B Venkatapur, A P S College of Engineering, Bangalore-78, Karnataka, India; V.D.Mytri, GND College of Engineering, Bidar, Karnataka, India; A.Damodaram, J N T U College of Engineering, Hyderabad, India

"Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than IJCIR must be honored. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee." © International Journal of Computing and ICT Research 2009. International Journal of Computing and ICT Research, ISSN 1818-1139 (Print), ISSN 1996-1065 (Online), Issue Vol.3, No.2 pp. 58-64, December 2009.

59


information being stored and published. Online portals providing image and video content like flickr.com, youtube.com, revver.com, etc. offer data in large amounts which creates the need for an efficient way of searching through the content. But the need for an efficient search and browsing method is not bound to those online portals. Archives of TV stations storing increasingly more digital content also need appropriate tools to find the material they are looking for. Additionally, with the already widely spread availability of digital acquisition devices (still image and video cameras) it is easy for everybody to acquire large amounts of digital data in short amounts of time.

Currently the search for specific content in such collections is mostly done through a query-by-text approach, exploiting manual annotation of the stored data. This approach suffers from a few drawbacks which arise from the nature of this method. First: manual annotation is a very time consuming process which second: might lead to a rather subjective result, depending on the person doing the annotation. Third: the result of the query depends highly on the quality of the annotations. Visual content that has not been transcribed into the metadata can therefore not be retrieved afterwards. Fourth: The result of a query can be manipulated by the type and number of tags associated with the visual data.

Our approach is to build up a similarity based structure of the multimedia database (here we focus on video as content) to overcome the need for an appropriate search example. Doing this, the user is enabled to browse through the database by picking an entity as starting point which represents the query most. During browsing the user is able to zoom in and out of the database content with variable step size representing the similarity of appearance. Entities showing up during the navigation, giving a better representation of the users query, provide the opportunity to narrow down the selection of possible matches.

With the development of multimedia retrieval/browsing systems quickly the problem of formulating a proper query arose. This lead to the definition of the so called semantic gap - a lack of coincidence between the information that one can extract from the visual data and the interpretation that the same data have for a user in a given situation [X. He et. al., 2004], [Smeulders A., 2000]. Different approaches have been examined to bridge this semantic gap [3-6 Smeulders A. et. al., 2000], [Broecker L. et. al., 2001], [Zhao R., Grosky W., 2002], [Zhao R., Grosky W., 2003] lead to multiple approaches towards user interface design i.e. the RotorBrowser [Dorai C., Venkatesh S., 2003] or VideoSOM [S. Yu., et. al, 2003].

When we talk about similarity based browsing, we focus on the visual content of the video. This means we want to allow the user to find temporally independent shots of video with similar content, which is different to cluster- temporal browsing where causal relations in storytelling are used for browsing [Barecke T., et. al, 2006]. We purely utilize content based similarity between different videos in the video database for browsing the database. A similar approach can be found in [ Rautiainen M. et. al, 2004] [Chen J., et. al, 1998], where the concept of a similarity pyramid is introduced for browsing large image databases. The idea of similarity pyramids was also applied to video databases in a system called ViBE [Chen J.Y., et. al].

Main objective of the present study is the direct use of a balanced tree structure for navigating through the database of keyframes, paired with an easy to use interface, offering a coarse to fine view on the grouped database content based on similarity.

2. BALANCED HIERARCHICAL CLUSTERING

To find structures of strong visual similarity we use unsupervised learning methods like clustering. In particular we are not only interested in the pure cluster partitioning, even more we want to capture the relationship between different clustering levels i.e. the clustering structure in the video database including their cluster and their subcluster partitions. To achieve this we use a standard agglomerative hierarchical clustering [Taskiran C., et. al, 2004], which runs through a series of partitions starting from singleton clusters containing a single image and terminating with a final cluster, containing all images of the database. This structure, usually represented as a dendogram, is post processed into a binary tree and used by our system for continuous browsing between different coarseness levels of similarity. 2.1. Feature Extraction Let the videos in the database be denoted as X1…… Xn. Every video is represented by a set of keyframes {xi1.......xim}, resulting in a total keyframe set { }kxx ,.....,1 for the entire video database. For every

60


keyframe { }ki xxx ,.....,1∈ a feature vector Di Rz ∈ is extracted.

A common way to extract keyframes for each video is to segment the data into shots and analyze the shots individually for representative keyframes. We use a divide-and-conquer approach that delivers multiple keyframes per shot in respect to its visual complexity. To achieve this, we compute MPEG-7 color layout descriptors [Johnson S.C., 1967] for each frame of the shot and fit a Gaussian mixture model to the feature set using k-means [Manjunath B., et. al, 2001] in combination with the Bayesian Information Criterion (BIC) for determining the number of clusters [Johnson S.C., 1967]. The entire procedure is illustrated in more detail in [Tamura H., et. al., 1978].

Fig. 1. A feature space representation of keyframes (a) leads to a hierarchical clustering (b). This clustering can be viewed as a dendogram with similarity measurement at y- axis (c) and will be postprocessed to a binary tree for our GUI (d)

The computation of the feature vector iz for every keyframe { }ki xxx ,.....,1∈ is based on our Baseline System definition [Manjunath B., et. al, 2001]. There we are using Color and Texture features with a equally weighted early fusion (i.e. concatenated). In particular, we use color histograms which are quantized to 888 ×× bins and also Tamura texture features [Tamura H., et. al., 1978] to form the feature vector. 2.2. Agglomerative Hierarchical Clustering The first step in using conventional agglomerative hierarchical clustering algorithm is the computation of a distance matrix ( )[ ]ji zzdD ,= where kji ,.....,1, = , with k the number of keyframes. The distance matrix D represents similarities between all pairs of keyframes and is used for successive fusion of clusters [Johnson S.C., 1967]. As distance function ( )ji zzd , serves the Euclidian distance.

The agglomerative hierarchical clustering creates a cluster { }ii xc = for each

keyframe { }ki xxx ,.....,1∈ , resulting in { }kccC ,.....,10 = disjoint singelton clusters, where CC ⊂0 . In this first step the distance matrix D is equal to the distances between the feature vectors of the keyframes. The cluster ci cj with the smallest distance ( )ji zzd , are fused together to form a new

cluster { }jik ccc ,= . After creating a new cluster, the distance matrix D has to be updated to represent the distance between the new cluster ck to each other cluster in C0\{ ck}. This recalculation of distances is usually done with one of the known linkage methods [Chen J.Y., et. al. 2000]. Considering the used linkage method the entire clustering structure will be more or less dilating i.e. individual elements not yet grouped are more likely to form new groups instead of being fused to existing groups. According to [Chen J.Y., et. al. 2000], the complete linkage method tends to be dilating and therefore resulting in more balanced dendograms compared to the single linkage method, which chains clusters together and therefore results in deep unbalanced dendograms. However, our goal is to use the resulting clustering structure for continuous similarity browsing of video databases. In order to achieve this, two points are important: First, the clustering must produce clusters with visual similar content and second, the dendrogram produced by the clustering has to be as balanced as possible. According to our experience, average linkage produces the visually most similar clusters. Unfortunately the clustering structure is not as balanced as desired for usable

61


browsing. Therefore a modification of the average linkage method was needed to achieve the desired properties.

Viewing the resulting dendogram of the clustering as a tree structure, this structure will hierarchically organize keyframes into groups of visual similar content, thereby retaining the relationship between different coarseness levels of similarity and tree depth. Let S denote the set of all tree nodes. Each node of the tree Ss∈ is associated with a set of keyframes 0CcCc ss ∉Λ⊂ representing the cluster of

keyframes. The number of elements in the cluster cs is denoted by sc . The children of a node Ss∈

denoted by ( ) Ssch ⊂ will partition the keyframes of the parent node so that

( ) rschrs cc∈∪=

The leaf nodes of the tree correspond to the extracted keyframes and are indexed by the set S0. Each leaf node contains a single keyframe, so for all 0Ssi ∈ we have { }ii xc = with 1=ic implying

that 00 CS = . This notation is derived from the notations of [Chen J.,et. al, 1998], [Chen J.Y.,et. al., 2000] [Taskiran C., et. al., 2004] . Furthermore we define ( )[ ]ji ccdD ,= as the updated distance matrix of distances between the pairwise different clusters ci and cj. The linkage method for calculating distances between clusters with

11 ff ji cc Λ is in our case the average linkage method enhanced by a weighted penalty, which

depends on the amount of elements in both clusters

( ) ( ) ( )∑ ∑∈ ∈

+∗+∗∗

=ii jjcx cx

jijiji

ji cczzdcc

ccd α,1,

We are naming this method: balanced linkage due to its tendency to form balanced trees with clusters of consistent visual properties. The weight can be set to 10 ≤≤α , which either results in having no effect to the average linkage or totally forcing the algorithm to produce balanced trees without any visual similarity. The chosen value was empirically evaluated and set to = 0:01. 2.3 Binary Tree Construction The usually chosen representation of hierarchical clustering is a dendogram, which illustrates the fusions made at each successive stage of analysis. In a dendogram, the elements being clustered are placed usually at the bottom of the diagram and show fusions of clusters through connecting lines. Another representation for hierarchical clustering is using sets showing the elements being clustered in their feature space. Such sets represent one particular step in the clustering process and may contain subsets illustrating previous clustering steps.

Because we want to use the clustering structure for navigation, we postprocess the dendogram into a binary tree. The binary tree structure enables us to efficiently follow a keyframe from the root of the tree, which contains the entire database, to the leaf of the tree, which only contains the selected keyframe. With every zoom in step, the system is presenting a more similar subtree considering the selected keyframe and leaving out the frames, which are less similar to the selected keyframe. The zoom out step, lets the system present a tree containing more dissimilar keyframes, which might be useful for refinement of the query. For the notation of the constructed binary tree, we refer to section 2.2. Let T = S be the binary tree, then the nodes Tni ∈ are the fusion points where clusters ji cc , are being fused together to form a new

cluster { }jik ccc ,= . An interesting outcome of the binary tree postprocessing is the creation of so called content

stripes. These structures represent clusters within the binary tree, in such a way that the keyframes of subclusters are ordered in stripes according to their similarity and therefore providing a more intuitive way of visualizing clusters at a particular level (Fig. 2). Content stripes replace the need to additionally compute spatial arrangements for keyframes within a cluster like done with similarity pyramides [Chen J., et. al. 1998], [Chen J.Y., et. al., 2000]. Therefore our method is not only able to construct a hierarchical database structure but also to build up a similarity based order within a cluster in one single step instead of

62


separating these tasks. Fig. 2. A binary tree representation of a sample clustering, transformed to a content stripe displaying similarity clusters in an one-dimensional order

3. GRAPHICAL USER INTERFACE

Retrieval systems based on keyframes and best-match similarity tend to present a localized view of the database to the user, rather then providing an overview of the entire database. For uses who do not clearly know what exactly they are searching for, it would be more efficient to let them browse through the database and allow them to dynamically redefine their search query.

In this section we would like to present the Navidgator graphical user interface, which allows a user to easily and efficiently browse a video database in respect to his selected query. In our system a browsing process is initialized with starting at the root of the hierarchical clustered binary tree. First, the user has to select his first keyframe out of a randomly sampled set from the entire database to formulate his query [X. He, et. al, 2004]. This keyframe will represent the visual concept, which will guide the user while browsing. The user is also able to dynamically refine his visual concept in every point during browsing by selecting another keyframe.

Browsing itself is performed by the given zooming tools. The user can either zoom-in or zoom-out in the database. A zoom-in action will narrow down the available keyframe according to his visual concept and a zoom-out action will display a coarser level of the database to the user. For better usability the interface provides a multi-level-zoom-in action and a multi-level-zoom-out action. Furthermore the user can perform a max-zoom-in action, which brings him straight to the most similar keyframes in the database or a max-zoom-out action, which brings him back to the root of the binary tree i.e. the top of the database. The depth of the database and the users current position are visualized by a vertical bar next to the zooming tools enabling an intuitive orientation. Additionally the user can utilize a click history to jump back to particular points of his browsing process. The Navidgator browsing interface is displayed in Fig. 3.

63


Fig. 3. The Navidgator browsing interface. The selected visual concept is displayed at the center. The lower area displays the cluster preview box, where the visual concept might be refined. Next to the selected keyframe the navigation tools are arranged

4. CONCLUSION

In this paper we have presented a new system for similarity-based browsing of multimedia databases. By using a balanced tree based on hierarchical clustering of the database content it is possible to supply users with an intuitive and easy- to-use browsing tool. We were able to improve search results by providing a set of navigation tools which support the decision tree like structure of the clustering. Because of our concept of an offline clustering and online retrieval we are able to efficiently perform a search on the entire database. The system offers coarse and detailed views on the database content with the opportunity to change the focus of search at any time. This enables the user to start navigation with a fuzzy visual concept and improve relevance incrementally while browsing.

5. REFERENCES

X. HE, D. CAI, J.R. WEN, W.Y. MA AND H.J. ZHANG, “ImageSeer: Clustering and Searching WWW Images Using Link and Page Layout Analysis”, Microsoft Technical Re port, (2004), MSR-TR-2004-38.

SMEULDERS A., WORRING M., SANTINI S., GUPTA A., JAIN R.: Content based image retrieval at the end of the early years. IEEE transactions on pattern analysis and machine intelligence, 22 (12) (2000), 1349-1379

BROECKER L., BOGEN M., CREMERS A.B.: Bridging the semantic gap in content- based image retrieval systems. In: Internet Multimedia Management Systems II, Volume 4519 of Presented at the Society of Photo-Optical Instrumentation Engineers (SPIE) Conference. (2001) 54-62

ZHAO R., GROSKY W.: Narrowing the semantic gap-improved text-based web document retrieval using visual features. Multimedia, IEEE Transactions on 4(2) (2002) 189-200

ZHAO R., GROSKY W.: Bridging the Semantic Gap in Image Retrieval. Distributed Multimedia Databases: Techniques and Applications (2003)

DORAI C., VENKATESH S., Bridging the semantic gap with computational media aesthetics. Multimedia, IEEE 10(2) (2003) 15-17

S. YU., D. CAI., J.R. WEN., W.Y. MA.,: “Improving pseudo-relevance feedback in web information retrieval using web page segmentation”, Proc. 12th World Wide Web Conference, (2003), Budapest, Hungary.

BARECKE T., KIJAK E., NURNBERGER A., DETYNIECKI M.: A som-based interface for video browsing. Image and Video Retrieval, Proceedings 4071 (2006) 506-509

RAUTIAINEN M., OJALA T., SEPPANEN T.: Cluster-temporal browsing of large news video databases.

64


IEEE Int. Conference on Multimedia and Expo 2 (2004) 751{754 CHEN J., BOUMAN, C., DALTON J.: Similarity pyramids for browsing and organization of large image

databases. SPIE Human Vision and Electronic Imaging III 3299 (1998) CHEN J.Y., BOUMAN C., DALTON J.: Hierarchical browsing and search of large image databases. Image

Processing, IEEE Transactions on 9, Issue 3 (March 2000) 442- 455 TASKIRAN C., CHEN J., ALBIOL A., TORRES L., BOUMAN C., DELP E.., A compressed video database

structured for active browsing and search, IEEE Transactions on Multimedia 6(1) (FEB 2004) 103{118

JOHNSON S.C.: Hierarchical clustering schemes. Psychometrika 32(3) (1967), 241-241 MANJUNATH B., OHM, J., VASUDEVAN, V., YAMADA, A.: Color and texture descriptors. IEEE Trans.

on Circuits Syst. for Video Techn. 11(6) (2001) MCQUEEN J.: Some methods for classification and analysis of multivariate observations. In Proceedings of

the Fifth Berkeley Symposium on Mathematical Statistics and Probability (1967) 281-297 TAMURA H., MORI S., Yamawaki, T.: Textual features corresponding to visual perception. IEEE

Transactions on Systems, Man, and Cybernetics SMC-8(6) (1978)

international journal of computing and ict research issue 2 2009-2.pdf · international journal of...

Documents