"The devil you know knows best": how online recommendations can benefit from social networking

The defining characteristics of the Internet today is an abundance of information and choice. Recommender Systems (RS), designed to alleviate this problem, have so far not been very successful, and recent research suggests that this is due to the lack of the social context and inter-personal trust. We simulated an online film RS with 60 participants, where recommender information was added to the recommendations, and a subset of these were attributed to friends of the participants. Participants overwhelmingly preferred recommendations from familiar recommenders with whom they shared interests and a high rating overlap. When recommenders were familiar, rating overlap was the most important decision factor, whereas when they were unfamiliar, the combination of profile similarity and rating overlap was important. We recommend that RS and social networking functionality should be integrated, and show how RS functionality can be added to an existing social networking system by visualising profile similarity.