Application of Ontology Techniques to View-Based Semantic Search and Browsing

We show how the benefits of the view-based search method, developed within the information retrieval community, can be extended with ontology-based search, developed within the Semantic Web community, and with semantic recommendations. As a proof of the concept, we have implemented an ontology- and view-based search engine and recommendation system Ontogator for RDF(S) repositories. Ontogator is innovative in two ways. Firstly, the RDFS-based ontologies used for annotating metadata are used in the user interface to facilitate view-based information retrieval. The views provide the user with an overview of the repository contents and a vocabulary for expressing search queries. Secondly, a semantic browsing function is provided by a recommender system. This system enriches instance level metadata by ontologies and provides the user with links to semantically related relevant resources. The semantic linkage is specified in terms of logical rules. To illustrate and discuss the ideas, a deployed application of Ontogator to a photo repository of the Helsinki University Museum is presented.