Exploitation of user profile in the information retrieval based on Bayesian approach

To satisfy the user information needs, the use of his/her profile is adopted in the information retrieval process. Our goal is to build and exploit the user profile in search process. To achieve this goal, we adopt a Bayesian network as a strategy for personalized search since it provides a clear formalism to model qualitatively and quantitatively information types, namely, a user query, documents, index terms, and user´s interests. In this paper, we present a Bayesian network model that captures the relationships between this information to evaluate the relevance as matching relationships between documents, user´s interest and query concepts. An experimental evaluation was designed to appraise the exploitation impact of the user profile defined by his/her interests on the search results relevance.