Mineração de Preferências Condicionais

Abstract

Content customization on queries to databases through the use of user's preferences is a research topic that has been raising a lot of interest within the database community in recent years. Such preferences are used for sorting and selecting the best tuples, those which most fulll the user's wishes. A topic of interest within this context is the elicitation of preferences, consisting of methods to enable the user to inform his choice on pairs of objects belonging to a database. This task may require a great eort from the user, depending on the size of the database, a fact that may discourage the use of the system by the user. In this work, we propose a methodology to infer user preferences in an automatic way through the task of mining, from a sample set of user preferences. The algorithm was designed to mine a special type of preferences called conditional preferences. The method we propose is based on the framework of Bayesian Belief Networks, a very ecient formalism for modeling situations involving uncertainty.