SOSEM: A Self-recommending Ontology Search Extension Mechanism Based on User Behavior

Abstract

Ontology search mechanism plays an important role in the discovery and usage of ontology. The current ontology search engines all adopt keyword-based search mechanism. It is difficult to clearly represent users’ search requirements because the number of the keywords input by a user is usually few. Extending the input keyword(s) by using WordNet will increase the recall rates and decrease the accuracy rates, because it enlarges the search scope imprecisely. To improve the performance of ontology searching, a self-recommending ontology search extension mechanism is proposed based on the study of the ontology search requirements and user search behaviors in this paper. The experimental results show that the proposed approach can effectively improve the matched coverage between query and target ontologies. Our method can also help user to conveniently search desired ontologies with high precision rates. Ontology search engines will get benefit from our method, because it will obviously improve the customer’s experience during ontology searching.