CSAIL Event Calendar: Previous Series
Context-aware Collaborative Filtering for Learning Interests of Mobile Users
The amount of information available to Internet users today is increasing rapidly, but users lack effective tools for obtaining accurate, up-to-date, and relevant information. This problem is especially acute for mobile users, who could greatly benefit from context-specific information but are constrained by the limited input, display, and processing capabilities of mobile devices. To address this problem, we built an automatic, personalized, context-aware event notification system for mobile users. In this system, a user agent automatically and unobtrusively learns user interests for the current context based on implicit rating and collaborative filtering. In this talk I describe the context-aware collaborative filtering algorithm used by this agent and present evaluation results comparing this approach to other approaches.