Social network and user context assisted personalization for recommender systems

In recommender systems, social networks are considered as a trusted source for user interests. In addition, user context can enhance users' decision making. In this paper, we design a new architecture for user personalization which combines both social network data and context data. Our system aggregates a user's preference data from various social networking services and then builds a centralized user profile which is accessible through public Web services. We also collect user's contextual information and store it in a central space which is also accessible through public Web services. Based on Service Oriented Architecture, recommender systems can flexibly utilize users' preference information and context to provide more desirable recommendations. We present how our system can integrate both types of data together and how they can be mapped in a meaningful way.