Mining User Contextual Preferences

User preferences  play an important role in database query personalization since they can be used for sorting and selecting  the objects which most fulfill the user wishes. In most situations user preferences are not static and may vary according to a multitude of user contexts.  Automatic tools for extracting contextual preferences without bothering the user are desirable. In this article, we propose CPrefMiner, a mining technique for mining user contextual preferences. We argue that  contextual preferences  can be naturally expressed by a Bayesian Preference Network (BPN). The method has been evaluated in a series of experiments executed on synthetic and real-world datasets and  proved to be  efficient to discover user contextual preferences.