Abstract
People use various Internet-based services including social networks to conduct tasks of diversified categories such as informational, professional, educational, hobbies, health, recreational, academic, and news. Many people not only maintain association with a number of different services but also share information with their social ties in their daily life. However, a person only consumes a subset of services and needs to communicate with a subset of social ties at any given context. In this paper, we present a framework called SenseFace that leverages sensory data coming from one's body sensor network and multimedia information contained within Internet-based services to recommend context-aware services and community of interest. We present the detailed design and implementation of the framework and share our preliminary test results.