HyKSS: A Multiple Ontology Approach to Hybrid Search

The rapid production of digital information has made the task of locating relevant information increasingly difficult. Keyword search alleviates this difficulty by retrieving documents containing keywords of interest, but suffers from issues such as ambiguity and synonymity. Semantic search resolves these issues, but is limited by the quality of semantic annotations that are likely to be imprecise and incomplete. In this proposal we introduce HyKSS, a Hybrid K eyword and Semantic Search system. HyKSS mitigates the weaknesses of these two approaches by employing both techniques in a hybrid ranking process in order to retrieve more relevant results. Further, HyKSS is not limited to queries over a single domain, but can employ multiple ontologies representing several domains from its customizable library in the ranking process. We expect to be able to show that the proposed hybrid system outperforms keyword search and semantic search.