The following information was submitted:
Transactions: WSEAS TRANSACTIONS ON COMPUTERS
Transactions ID Number: 28-202
Full Name: Saravanan Muthaiyah
Position: Senior Lecturer
Age: ON
Sex: Male
Address: Multimedia University, Faculty of Management, 63100 Cyberjaya, Selangor, Malaysia
Country: MALAYSIA
Tel: 019-3152744
Tel prefix:
Fax: 03-83125590
E-mail address: smuthaiy@gmu.edu
Other E-mails: saravanan.muthaiyah@mmu.edu.my
Title of the Paper: Managing Ontology Change and Evolution via a Hybrid Matching Algorithm
Authors as they appear in the Paper: Saravanan Muthaiyah, Marcel Barbulescu, Larry Kerschberg
Email addresses of all the authors: smuthaiy@gmu.edu,mbarbulescu@gmu.edu,kersch@gmu.edu
Number of paper pages: 11
Abstract: In this paper, we present the problem of ontology evolution and change management. We provide a systematic approach to solve the problem by adopting a multi-agent system (MAS). The core of our solution is the Semantic Relatedness Score (SRS) which is an aggregate score of five well-tested semantic as well as syntactic algorithms. The focus of this paper is to resolve current problems related to ontology upgrade and managing evolution amongst shared ontologies. This paper highlights issues pertaining to ontological changes in a shared ontology environment which includes creating, renaming, deletion and modification of existing classes. These changes will definitely impact shared concepts and users would have to update their local ontologies to be consistent with changes in the commonly shared ontology. We propose a less laborious method to achieve this by using a semi-automated approach where a bulk of the processing is carried out by matching agents that would elim!
inate extraneous data and hence would only recommend to the ontologist data that can actually be upgraded. We have also designed and built a prototype in the Java Agent DEvelopment Framework (JADE) for proof-of-concept.
Keywords: Hierarchical Repository, Semantic Matching, Syntactic Matching, Agent, Ontology
EXTENSION of the file: .doc
Special (Invited) Session: A Hybrid Similarity Matching Algorithm for Mapping and Upgrading Ontologies via a Multi-Agent System
Organizer of the Session: 591-378
How Did you learn about congress:
IP ADDRESS: 60.49.115.80
