OntoSmart: Um modelo de recuperação de informação baseado em ontologia

Abstract

Information retrieval system is a mediator element between a stock of information and its users. Its effectiveness depends on representation language of information items and requests of its users. This work presents an ontology-based information retrieval model which uses the formal structure of Vector Space Model. The vector that represent a document is created during the automatic indexing process, in which an ontology provide new terms in order to semantically enrich that representation. The search vector is created from a query expansion process, in which new terms are added in the search expression initially formulated by the user from inferences in the ontology. Using the proposed model, the OntoSmart system is being developed. Preliminary results show a significant improvement in the precision of search results.