Authors:
Yin-Fu Huang
and
Yu-Yu Huang
Affiliation:
Graduate School of Computer Science and Information Engineering, National Yunlin University of Science and Technology, Taiwan
Keyword(s):
Ontology construction, ontology learning, Is-A and Parts-of relationships, word correlation, hierarchical clustering.
Related
Ontology
Subjects/Areas/Topics:
Ontology and the Semantic Web
;
Web Information Systems and Technologies
;
Web Interfaces and Applications
Abstract:
This paper proposed a general framework that could automatically construct domain ontology on a collection of documents with the help of The Free Dictionary, WordNet, and Wikipedia Categories. Both explicit and implicit features of index terms in documents are used to evaluate word correlations and then to construct Is-A relationships in the framework. Thus, the built ontology would consist of 1) concepts, 2) Is-A and Parts-of relationships among concepts, and 3) word relationships. Besides, the built ontology could be further refined by learning from incremental documents periodically. To help users browse the built ontology, an ontology browsing system was implemented and provided different search modes and functionality to facilitate searching a variety of relationships.