Improved Reinforcement-Based Profile Learning for Documents Filtering

today the problem is not the availability of the information but how to get the related information. A personalized information filtering system must be able to tailor to current interests of the user and to adapt as they change over time. This research has proposed a content-based personal information system that learns the user preferences by analyzing the content of the document and building the user profile. The proposed filtering system monitors a stream of incoming documents to deliver only those matches the user profiles. This system is called RePLS; an agent-based Reinforcement Profile Learning System with adaptive information filtering. The agent approach is used because of its autonomous and adaptive capabilities to perform the filtering. The core of this system is an improved term weighting method which is called “Purity term weighting” to measure the importance of the most suitable terms represented in each profile. The top selected terms are then used to filter the incoming documents to the learned user profiles.