relational database schema to ontology mapping approaches zand-mogha… · references relational...

RELATIONAL DATABASE SCHEMA TO ONTOLOGY MAPPING APPROACHES

by Yassaman Zand Moghaddam and Joe D. Horton

November 10th

2010

Outline

A brief description of mapping and correspondenceCorrespondence PatternsDatabase to ontology mapping process

A naïve solutionRelational database to ontology mapping approachesOntology matchingMatching evaluation parametersThe ideaEvaluation methodAn overview of the MapOnto tool

Relational Database Schema to Ontology Mapping Approaches

2

2

What do “Mapping”

and “Correspondence”

mean?


3

Mapping is the process of finding relationships or correspondences between entities.Correspondence is the relation holding or supposed to hold according to a particular mapping algorithm between entities.Each correspondence set a bridge between entities.

3

Correspondence Patterns


4

Correspondence patterns (Scharffe & Fensel, 2008) are like templates to model ontology alignments (The Alignment is the output of ontology mapping). They capture regularities recurring when aligning ontologies.In many applications and domains, specific correspondence patterns exist. Relational database to ontology mapping is one of those applications.

4

Correspondence patterns identified in relational database to ontology mapping application:


5

Direct MappingJoin/Union MappingProjectionSelectionValue transformationCombinations

5

Database to Ontology Mapping Process


6

At a conceptual level a database and an ontology are semantically related and correspondences are established between the database components and the ontology components.

6

Database to Ontology Mapping Approaches


7

Figure1. Classification of database to ontology mapping approaches (Ghawi & Cullot, 2007)

7

Discovering mappings between a relational database schema and an ontology usually has two phases:


8

Searching for simple mappings between entities in the relational database schema and the ontology.

Constructing complex compositions based on simple mappings.

8

A naïve solution


9

Find the “shortest” connection between concepts.It involves:

Finding the minimum spanning tree(s) connecting the corresponded conceptsEncoding the tree(s) into formulas

9

An example of the naïve solution (An, Borgida, & Mylopoulos, 2005)


10

Figure 2. Relational table, Ontology, and Correspondences (An, Borgida, & Mylopoulos, 2006).

10

An example of the naïve solution


11

Simple correspondences:T : Employee.ssn O : Employee.hasSsnT : Employee.name O : Employee.hasNameT : Employee.dept O : Department.hasDeptNumberT : Employee.proj O : Worksite.hasNumber

Expected formula:T :Employee(ssn,name,dept,proj) :-O:Employee(x1), O:hasSsn(x1,ssn), O:hasName(x1,name), O:Department(x2), O:works_for(x1,x2), O:hasDeptNumber(x2,dept), O:Worksite(x3),

O:works_on(x1,x3), O:hasNumber(x3,proj).

11



12

In some cases, because of the known relational schema design rules, the spanning tree may not provide the desired semantics for a table.

Project (name, manager ) -> “manager” is functionally dependent on “name”

T : Project.name O : Worksite.hasName

T : Project.manager O : Employee.hasSsn

12

WorksiteEmployee works _on

1..*

0..1



13

We need to find a functional connection from “Worksite”

to “Employee”

Project (name, manager )

WorksiteEmployee

controls

1..10..1Department1..1 0..*

manages

Mapping approaches


14

Some of the approaches that address the problem of mapping databases to ontologies are:

RDB2Onto (Laclav’k, 2007)DB2OWL (Cullot, Ghawi, & Ytongnon, 2007)Relational OWL (Prez & Conrad, 2005)VisAVis (Konstantinou, Spanos, Chalas, Solidakis, & Mitrou, 2006)Discovering Simple Mappings Between Relational Database Schema and Ontologies (Hu & Qu, 2007)

14

Mapping approaches


15

Another Literature review on the topic “Database to Ontology Mapping” in (Sataya S. et al., 2009), classifies the reviewed approaches based on:

Mapping CreationMapping Representation and AccessibilityMapping ImplementationQuery ImplementationApplication DomainData Integration

15

Ontology matching Figure2.

Classification of ontology matching techniques (Euzenat & Shvaiko, 2007)


16

16

Ontology matching Figure2. Classification of ontology matching techniques (Euzenat & Shvaiko, 2007)


17

17

Ontology matching Figure3. Extracted from Classification of ontology matching techniques (Euzenat &

Shvaiko, 2007)


18

18

Matching evaluation parameters


19

Precision =

Recall =

F-Measure = 0≤α≤1

Overall =

1: True Correspondences2: Derived CorrespondencesA: False Negative

C: False PositiveB: True Positive

D: True Negative

19

1 2

DA B C

The idea


20

Inputs:

Outputs:

20

RDB Schema

Ontology

Intermediate Graph

Intermediate GraphMatching

(Structure- based)

A set of correspondences Some

string- based checks

A refined set of correspondences

Graph isomorphism: An example (source: wikipedia)


21

An isomorphism of graphs G and H is a bijection between the vertex sets of G and H such that any two vertices u and v of G are adjacent in G if and only if f(u) and f(v) are adjacent in H.

21

G:

H:

f(a) = 1f(b) = 6f(c) = 8f(d) = 3f(g) = 5f(h) = 2f(i) = 4f(j) = 7

An isomorphism between G and H:

f: V(G) V(H)

String-based Check


22

22

String-based techniques are often used in order to match names and name descriptions of entities. These techniques consider strings as sequences of letters in an alphabet. In general, they are based on the intuition: the more similar the strings, the more likely they are to represent the same concepts. Usually, distance functions map a pair of strings to a real number, where a smaller value of the real number indicates a greater similarity between the strings. Examples: prefix, suffix, edit, and n-gram distances.

Evaluation Method


23

One method can be the repetition of the process in reverse which means:construction of the relational database schema and the ontology from intermediate graphscomparison of results with original ones (inputs).

If they are equivalent, there is a high probability that the algorithm can work correctly on any inputs.

23

MapOnto Tool (An, Borgida, & Mylopoulos, 2006 )

A system for constructing complex mappings between ontologies and relational or XML database schemas.


24

The inputs of the tool are:An ontology specified in an ontology representation language like OWLA relational or XML database schemaA set of simple correspondences

The output of the system is:An ordered list of complex mapping formulas expressed in Horn clauses.

The issue is:Simple mappings between the relational database schema and the ontology are specified manually.

24

MapOnto Evaluation


25

Datasets that were used in evaluation of the MapOnto tool are in different real world domains and were implemented independently.

Table1. Characteristics of Schemas and ontologies for the experiments (An, Borgida, & Mylopoulos, 2006).

25


26

Thank You

26

References


27

[1] Yuan An, Alexander Borgida, and John Mylopoulos, Constructing Complex Semantic Mappings Between (XML) Data and Ontologies, International Semantic Web Conference, 2005, pp. 6-20.

[2] -, Inferring Complex Semantic Mappings Between Relational Tables and Ontologies

from Simple Correspondences, (OTM) Conferences (2), 2005, pp. 1152-1169.

[3] Tim Berners-Lee, James Hendler, and Ora

Lassila, The SemanticWeb, vol. 284, 2001, pp. 34-43.

[4] C. Blakeley, RDF Views of SQL Data (Declarative SQL Schema to RDF Mapping), Open Link Software, 2007.

[5] Kevin Chen-Chuan

Chang, Bin He, Chengkai

Li, Mitesh

Patel, and Zhen Zhang, Structured Databases on the Web:Observations

and Implications, vol.33, 2004, pp.61-70.

[6] Huajun

Chen, Yimin

Wang, Heng

Wang, Yuxin

Mao, Jinmin

Tang, Cunyin

Zhou, Ainin

Yin, and Zhaohui

Wu, Towards a Semantic Web of Relational Databases:A

Practical Semantic Toolkit and an In-use Case from Traditional Chinese Medicine, 5th

International Semantic Web Conference, Springer, 2006, pp.750-763.

27

References


28

[7] Nadine Cullot, Raji

Ghawi, and Kokou

Ytongnon, Db2owl: A Tool for Automatic Database-to-Ontology Mapping, SEBD, 2007, pp.491-494.

[8] Hong-Hai

Do, Sergey Melnik, and Erhard Rahm, Comparison of Schema Matching Evaluations, In Proceedings of the 2nd International Workshop on Web Databases, 2002, pp. 221-237.

[9] Jerome Euzenat

and Pavel

Shvaiko, Ontology Matching, Springer-

Verlag,Berlin, Heidelberg, 2007.

[10] R. Ghawi

and N.Cullot, Database-to-Ontology Mapping Generation for Semantic Interoperability, 2007.

[11] Thomas R. Gruber, A Translation Approach to Portable Ontology Specications, Tech. report, Stanford Knowledge Systems Laboratory, 1993.

[12] Mauricio Hernandez, Lucian Popa, Howard Ho, and Felix Naumann, Clio: A Schema Mapping Tool for Information Integration, IEEE Computer Society, 2005.

28

References


29

[13] Wei

Hu

and Yuzhong

Qu, Discovering Simple Mappings Between Relational Database Schemas and Ontologies, 6th

International Semantic Web Conference and 2nd

Asian Semantic Web Conference, 2007, pp. 225-238.

[14] Nikolaos

Konstantinou, Dimitriosemmanuel

Spanos, Michael Chalas, Emmanuel Solidakis, and Nikolas

Mitrou, Visavis: An Approach to an Intermediate Layer between Ontologies

and Relational Database Contents, International Workshop on Web Information Systems Modeling (WISM2006), 2006.

[15] Michael Laclav'k, Tech. report, Institute of Informatics, Slovak Academy of Science, 2007.

[16] Eduard

Dragut

Ramon Lawrence, Composing Mappings Between Schemas Using a Reference Ontology, International Conference on Ontologies, Databases and Application of Semantics (ODBASE), Springer, 2004, pp.783-800.

[17] Igor Myroshnichenko

and Marguerite C. Murphy, Mapping (ER) Schemasto

(OWL) Ontologies, International Conference on Semantic Computing, 2009, pp. 324-329.

[18] Natalya

F. Noy

and Deborah L. McGuinness, Ontology Development 101: a Guide to Creating Your First Ontology, Tech. report, Stanford Knowledge Systems Laboratory, 2002.

29

References


30

[19] Ontology, http://semanticweb.org/wiki/ontology, 2010.

[20] P. Papapanagiotou, P. Katsiouli, V. Tsetsos, C. Anagnostopoulos, and S. Hadjiefthymiades, Ronto: Relational to Ontology Schema Matching, AIS SIGSEMIS BULLETIN 3 (3-4), 2006, pp. 32-36.

[21] Cristian

Prez, and Laborda

Stefan Conrad, Relational.Owl-

a Data and Schema Representation Format Based on Owl, APCCM05: Proceedings of the 2nd

Asia-Pacic

conference on Conceptual modelling, Australian Computer Society Inc., 2005, pp.89-96.

[22] MapOnto

Project, http://www.cs.toronto.edu/semanticweb/maponto/index.html, 2005.

[23] Samantha Rajapaksha, Relational Database Schema to Ontology Mapping Tool, Tech. report, Sri Lanka Institute of Information Technology, 2008.

[24] Sataya

S. Sahoo, Wolfang

Halb, Sebastian Hellmann, Kingsley Idehen, Ted Thibodeau

Jr, Soren

Auer, Juan Sequeda, and Ahmed Ezzat, A Survey of Current Approaches for Mapping of Relational Databases to RDF, Tech. report, W3C IncubatorGroup, 2009.

[25] Franςois

Scharffe

and Dieter Fensel, Correspondence Patterns for Ontology Alignment, Proceedings of the 16th international conference on Knowledge Engineering: Practice

and Patterns, EKAW '08, pp. 83-92, 2008

30

http://semanticweb.org/wiki/ontology

http://www.cs.toronto.edu/semanticweb/maponto/index.html

References


31

[26] Swoogle, http://en.wikipedia.org/wiki/swoogle,2010.

[27] Mike Uschold

and Michael Gruninger, Ontologies: Principles, Methods and Applications, vol.11, 1996.

[28] Zhuoming

Xu, Shichao

Zhang, and Yisheng

Dong, Mapping between Relational Database Schema and Owl Ontology for Deep Annotation, WI'06: Proceedings of the 2006 IEEE/WIC/ACM InternationalConferenceon

Web Intelligence, IEEE Computer Society, 2006, pp. 548-552.

31

http://en.wikipedia.org/wiki/swoogle,2010

relational database schema to ontology mapping approaches zand-mogha… · references relational...

Documents