Generating Predictive Movie Recommendations from Trust in Social Networks

Abstract

Social networks are growing in number and size, with hundreds of millions of user accounts among them. One added benefit of these networks is that they allow users to encode more information about their relationships than just stating who they know. In this work, we are particularly interested in trust relationships, and how they can be used in designing interfaces. In this paper, we present FilmTrust, a website that uses trust in webbased social networks to create predictive movie recommendations. Using the FilmTrust system as a foundation, we show that these recommendations are more accurate than other techniques when the user's opinions about a film are divergent from the average. We discuss this technique both as an application of social network analysis, as well as how it suggests other analyses that can be performed to help improve collaborative filtering algorithms of all types. Report Documentation Page Form Approved OMB No. 0704-0188Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number.

Development of a group recommender application in a Social Network

“…There are specific approaches that use a custom trust network to recommend items. One example is FilmTrust [21], which exploits a custom network of trust among users according to movie preferences. However, these specific trust networks are quite difficult to generate because they require explicit feedback from users, and these can generate rejection.…”

“…Besides, there is recent work reporting significant recommendation performance improvement for social recommender systems [21,34,35,36,37,38,39]. On the other hand, there are also unsuccessful attempts at applying social recommendation [40,41].…”

“…Within this environment, we are able to infer much of the information needed to perform Social Group Recommendations directly from Social Networks [21,43]. Previously, the acquisition of such social data had to be performed by means of tedious questionnaires.…”

“…The use of trust and other social knowledge obtained from Social Networks in the development of recommender systems is not new [21,55]. We have reviewed several existing works [56,19] that identify the factors to be analysed.…”

“…For a given user u in group G a , we take into account the predicted ratings,r v,i , for the rest of the group members, v ∈ G a , v u, weighted by the trust between the two users, t u,v . This follows [21], where a method for group recommendations using trust is proposed.…”