Mineração de Preferências Condicionais

Nádia Félix Felipe da SilvaSandra de Amo

A lot of research involving formalisms to specify preferences, query languages for databases supporting preferences and techniques for eliciting and inferring preferences are emerging with the goal of filtering the answer of queries, by finding those answer which most fulfill the user’s desire or preferences. Elicitation and deduction are characterized by means of inferring the user’s choices with the least effort. However, depending on the size of the database, this task may be unfeasible to be achieved manually. In this paper we propose the algorithm CPrefMiner, an automatic tool for inference of user preferences. CPrefMiner is designed to mine a special kind of preferences, the so called conditional preferences. The method we propose is based on the framework of Bayesian Belief Networks, a very efficient formalism for modeling situations involving uncertainty.

Caso o link acima esteja inválido, faça uma busca pelo texto completo na Web: Buscar na Web

Biblioteca Digital Brasileira de Computação - Contato:
     Mantida por: