Please use this identifier to cite or link to this item:
|Title:||On the Merging of Domain-Specific Heterogeneous Ontologies Using WordNet and Web Pattern-Based Queries|
|Keywords:||Knowledge management and representation; ontology merging; semantic and web-based statistical techniques; WordNet; precision/recall experimental evaluation|
|Publisher:||Journal of Information & Knowledge Management|
|Abstract:||Ontologies are a basic interest in various computer science disciplines such as semantic web, information retrieval, database design, etc. They aim at providing a formal, explicit and shared conceptualization and understanding of common domains between different communities. In addition, they allow for concepts and their constraints of a specific domain to be explicitly defined. However, the distributed nature of ontology development and the differences in viewpoints of ontology engineers have resulted in the so-called "semantic heterogeneity" between ontologies. Semantic heterogeneity constitutes the major obstacle against achieving interoperability between ontologies. To overcome this obstacle, we present a multipurpose framework which exploits the WordNet generic knowledge base for the following purposes: (i) discovering and correcting the incorrect semantic relations between the concepts of the ontology in a specific domain, this step is a primary step of ontology merging; (ii) merging domain-specific ontologies through computing semantic relations between their concepts; (iii) handling the issue of missing concepts in WordNet through the acquisition of statistical information on the Web; (iv) enriching WordNet with these missing concepts. An experimental instantiation of the framework and comparisons with state-of-the-art syntactic- and semantic-based systems validate our proposal.|
|Appears in Collections:||Faculty & Staff Scientific Research publications|
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.