Weighted ontology-based search exploiting semantic similarity

  • Authors:
  • Kuo Zhang;Jie Tang;MingCai Hong;JuanZi Li;Wei Wei

  • Affiliations:
  • Knowledge Engineering Lab, Department of Computer Science, Tsinghua University, Beijing, P.R. China;Knowledge Engineering Lab, Department of Computer Science, Tsinghua University, Beijing, P.R. China;Knowledge Engineering Lab, Department of Computer Science, Tsinghua University, Beijing, P.R. China;Knowledge Engineering Lab, Department of Computer Science, Tsinghua University, Beijing, P.R. China;Software Engineering School, Xi’an Jiaotong University, Xi’an, P.R. China

  • Venue:
  • APWeb'06 Proceedings of the 8th Asia-Pacific Web conference on Frontiers of WWW Research and Development
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper is concerned with the problem of semantic search. By semantic search, we mean searching for instances from knowledge base. Given a query, we are to retrieve ‘relevant’ instances, including those that contain the query keywords and those that do not contain the keywords. This is contrast to the traditional approaches of generating a ranked list of documents that contain the keywords. Specifically, we first employ keyword based search method to retrieve instances for a query; then a proposed method of semantic feedback is performed to refine the search results; and then we conduct re-retrieval by making use of relations and instance similarities. To make the search more effective, we use weighted ontology as the underlying data model in which importances are assigned to different concepts and relations. As far as we know, exploiting instance similarities in search on weighted ontology has not been investigated previously. For the task of instance similarity calculation, we exploit both concept hierarchy and properties. We applied our methods to a software domain. Empirical evaluation indicates that the proposed methods can improve the search performance significantly.