intelligent information system laboratory1375ae4d4dc1051c3bfb… · intelligent information system...
TRANSCRIPT
Intelligent Information System Laboratory
1
INTRODUCTIONS
RESEARCH TOPICS
ONGOING PAPERS
AND
RESEARCH CHALLENGES
Intelligent Information System Lab. Korea University
1 . PROFESSOR IN -JEONG CHUNG
2 . STUDENTS OF I IS LAB .
Introduction to IIS Lab.2
2 . STUDENTS OF I IS LAB .
Intelligent Information System Lab. Korea University
ProfessorIn-Jeong, Chung
Dept. of Computer and Information Science
Korea University
http://[email protected]
3
Lectures- Data structure- Artificial intelligence- Automata
Intelligent Information System Lab. Korea University
Students of IIS Lab.
� Doctor course
� Sohn jong-soo
� Master course
� Kim doh-hyug (3)
Wang Qing (1)
4
� Wang Qing (1)
� Pei Yunfeng (1)
� Researcher
� Kwon Kyunglak
Intelligent Information System Lab. Korea University
SEMANTIC WEB
Research Topics5
INTELLIGENT WEB SERVICES
SOCIAL NETWORK
COLLECTIVE INTELLIGENCE
Intelligent Information System Lab. Korea University
Semantic Web
� Goal of semantic web
� Computers as well as people can find, read, understand, and use data over the World Wide Web to accomplish useful goals for users
� Web VS. Semantic web
6
� Web VS. Semantic web
� Web
�HTML
� Hyper link
� Semantic web
� Ontology languages (XML, RDF, Topic Map, OWL… etc.)
� Semantic annotation (semantic meta-data)
Intelligent Information System Lab. Korea University
Subclass of Subclass of
Subclass of
Equivalent Equivalent
XML
RDF
DAML
OIL
DAML+OIL
OWL
8
Ontology : An ontology is a formal specification of a shared conceptualization
Topic Map
And its development tools
Intelligent Information System Lab. Korea University
RDF and RDF-S
� RDF
� Describing all sorts of information and meta data
� Described Semantics by RDF
� Structure of triple
� Contains resource property and value of property
9
� Contains resource property and value of property
� Every RDF element should be defined by URIs.
� RDF-S
� Framework for constructingontologies
� Providing means to specifybasic vocabularies
Intelligent Information System Lab. Korea University
OWL
� A kind of ontology language for semantic web
� OWL : Web Ontology Language
� More vocabulary for describing properties and classes
� Between classes, cardinality, equality, richer typing of properties, characteristics of properties, and enumerated classes
10
characteristics of properties, and enumerated classes
� Related projects
� DBpedia
� FOAF
� Linking Open Data
� NextBio
Intelligent Information System Lab. Korea University
SEMANTIC WEB
INTELLIGENT WEB SERVICES
Research Topics11
INTELLIGENT WEB SERVICES
SOCIAL NETWORK
COLLECTIVE INTELLIGENCE
Intelligent Information System Lab. Korea University
Intelligent web services
� Web services � Programmable application logic accessible
� Using standard internet protocols
� Combine the best aspects of component-based development and Web
� Block-box functionality for the reuse
� Without worrying about how the service is implemented
12
� Without worrying about how the service is implemented
� Accessible via ubiquitous web protocols
� Such as HTTP and data format such as XML
� Features� Application components and programmable application
� Self-contained, Self-describing components
� Accessible as components communicating with open protocols
Intelligent Information System Lab. Korea University
Web Service --- Web
UDDI --- URI
WSDL --- HTML
SOAP --- HTTP
Intelligent Web servicesOWL-S
13
Web services
OWL-SEBPL4WSDAML-S
ProtocolsSOAPXML-RPCRESTLDAP
Intelligent Information System Lab. Korea University
Semantic web services15
wide variety ofagent technologies forautomated Web servicediscovery, execution,composition, and
Intelligent Information System Lab. Korea University
composition, andinteroperation
SEMANTIC WEB
INTELLIGENT WEB SERVICES
Research Topics16
INTELLIGENT WEB SERVICES
SOCIAL NETWORK
COLLECTIVE INTELLIGENCE
Intelligent Information System Lab. Korea University
Social network analysis
� Social network analysis
� Mapping and measuring of relationships and flows
�Between people, groups, organizations, computers, URLs, and other connected information/knowledge entities
� The nodes in the network
17
� The nodes in the network
� People and groups while the links
� Showing relationships or flowsbetween the nodes
Intelligent Information System Lab. Korea University
Measures in social network analysis
� Betweenness
� The extent to which a node lies between other nodes
� Bridge
� An edge is said to be a bridge
Centrality
18
� Centrality
� Giving a rough indication of the social power of a node
� Closeness
� Near all other individuals in a network
� Density
� The degree a respondent's ties know one another/ proportion of ties among an individual's nominees
Intelligent Information System Lab. Korea University
Betweenness
Bridge
Centrality
Closeness
Density
19
Measures in social network analysis
And other measures
Intelligent Information System Lab. Korea University
Social network service
� Social network analysis != Social network service
� Social network service (SNS)
�Making relation service for people to people on the web
� Major social network service
� Twitter, Facebook, Cyworld, Me2Day and etc.
20
� Twitter, Facebook, Cyworld, Me2Day and etc.
Intelligent Information System Lab. Korea University
SEMANTIC WEB
INTELLIGENT WEB SERVICES
Research Topics21
INTELLIGENT WEB SERVICES
SOCIAL NETWORK
COLLECTIVE INTELLIGENCE
Intelligent Information System Lab. Korea University
Collective intelligence
� Definition
� Shared or group intelligence that emerges from collaboration and competition of many individuals
� Four principles
� Openness
22
� Openness
� Peering
� Sharing
� Acting globally
Intelligent Information System Lab. Korea University
Collective intelligence
� Concept of collective intelligence
� Combining of behavior, preferences, or ideas of a group of people to create novel insights
� Shared or group intelligence
� Emerges from the collaboration and competition of many
23
� Emerges from the collaboration and competition of many individuals
� Example
� Wikipedia:
� Online encyclopedia created entirely from user contributions
� Google page ranking:
�How many other pages link to the page
Intelligent Information System Lab. Korea University
Three components
� Three components harnessing collective intelligence
� User interaction
� Aggregating by users
� Personalized contents
� User interaction data
24
� Aggregated data
Intelligent Information System Lab. Korea University
Benefits of collective intelligence
� Higher retention rates
� More users interact with the application
� Greater opportunities to market to the user
� Greater the number of interactions
Greater the number of pages visited by the user
25
� Greater the number of pages visited by the user
� Higher probability of a user completing a transaction and finding information of interest
� More contextually relevant information that a user finds
� Boosting search engine rankings
� More users participate and contribute content
� More content available in application
Intelligent Information System Lab. Korea University
Working papers
� FOAF management using OLAP system
� User profile description language : FOAF
� Written by OWL
� Adding RSS data
� Using OLAP system
27
� Using OLAP system
� Multi-dimensional analysis
� Ontology generation using CI
� Very difficult research topic : Ontology generation
� Proposes user participation based ontology generation method
Intelligent Information System Lab. Korea University
Working papers
� Social network management
� Ontology based Mp3 metadata writing
� MP3 metadata management system
� Using semantic web tech.
Intelligent process control with RFID
28
� Intelligent process control with RFID
� RFID based intelligence process control
� Proposes process tree
Intelligent Information System Lab. Korea University
SEMANTIC WEB
SOCIAL NETWORK MANAGEMENT
How to adapt our interests to network management domain?
29
SOCIAL NETWORK MANAGEMENT
COLLECTIVE INTELLIGENCE
Intelligent Information System Lab. Korea University
Semantic web and network management
� Ontology-Based Network Management
� Many research projects in a number of different network management and security scenarios
� Semantic management: application of ontologies for the integration of management information models (2003)
30
integration of management information models (2003)
�Benefits of using ontologies in the management of high speed networks (2004)
� Security policy instantiation to reactto network attacks–An ontology-basedapproach using OWL and SWRL(2008)
Intelligent Information System Lab. Korea University
Semantic web and smart grid
� Multi-agent system in distributed smart grid
� Rule description and inference
�Multi-Agent Systems in a DistributedSmart Grid: Design andImplementation (2009)
31
� Decision making
� Design of ontology-based decisionsupport software system for grid dispatching (2004)
Intelligent Information System Lab. Korea University
Adapting collective intelligence
Algorithm Application publication
Bee hive Algo. Routing in networks Wedde et al (2004~7)
Bee hive Algo. Qos unicast routing scheme Wang et al (2007)
Swan Network management of IP networks Gupta and Koul (2007)
Bee system TSP problems Lucic and Teodorovic (2003)
32
Bee system TSP problems Lucic and Teodorovic (2003)
Bee colony optimization Routing and wavelength assignment Markovic et al (2007)
BeeAdhoc Routing in mobile ad hoc networks Wedde and Farooq (2005)
Artificial bee colony Network reconfiguration problem Rao et al (2008)
Honeybee search strategies
Routing and congestion avoidancein internet services
Walker (2004)
BeeAIS Security in the challenging MANET Saleem and Farooq (2007)
Intelligent Information System Lab. Korea University
Adapting social network analysis
� Change point of view!
� Human �computers, swichs
� Social network � computer network
� Then, we can deduct new type of network monitoring system
33
system
� Using measurement of SNA
� Closeness, centrality, density, bridge …
� Quick computation and simple
� Machine learning step
� Do not required
Intelligent Information System Lab. Korea University