Using social Trust for Recommender Systems Jennifer golbeck Human-Computer Interaction Lab University of Maryland, College Park How many cows in Texas? W ABDUCTIONS TO DAT 0.527,226 http://www.cowabduction.com
1 Using Social Trust for Recommender Systems Jennifer Golbeck Human-Computer Interaction Lab University of Maryland, College Park 2/39 How many cows in Texas? http://www.cowabduction.com/
facebook LIVEJOURNAL ywO Linked in EzorpIa.com myspace. C 圆■■■■ Xanga mixis >1.0000gye hi5 orkus YAHoOL 360 Adult riendfinderco China8en肉 mULTiPLyOfriendster You Tube
2 3/39 4/39 >1,000,000,000
Introduction a Trust is a way of doing social personalization a Understand how to compute trust a See applications where trust is being Used for creating recommendations 5/39 Defining trust a Overall, trust is very complex Involves personal background, history of interaction, context, similarity, reputation, etc a Sociological definitions a Trust requires a belief and a commitment E.g. Bob believes Frank will provide reliable information thus Bob is willing to act on that information Similar to a bet a In the context of recommender systems, trust is generally used to describe similarity in opInion Ignores authority, correctness on facts
3 5/39 Introduction Trust is a way of doing social personalization Understand how to compute trust See applications where trust is being used for creating recommendations 6/39 Defining Trust Overall, trust is very complex Involves personal background, history of interaction, context, similarity, reputation, etc. Sociological definitions Trust requires a belief and a commitment E.g. Bob believes Frank will provide reliable information thus Bob is willing to act on that information Similar to a bet In the context of recommender systems, trust is generally used to describe similarity in opinion Ignores authority, correctness on facts
Trust inference The Goal: Select two individuals-the source(node a) and sink (node c)-and recommend to the source how much to trust the sink B
4 7/39 8/39 Trust Inference The Goal: Select two individuals - the source (node A) and sink (node C) - and recommend to the source how much to trust the sink. A B C tAB tBC tAC
Major Algorithms-Networks u Advogato(Levien Appleseed(Ziegler and Lausen a MoleTrust(Massa and Avesani TidalTrust(Golbeck) Advogato Attack resistant Maximum network flow based on Ford-Fulkerson Node capacities determined by the distance from the This is a single source, multiple sink problem with capacities on the nodes u Network flow works on a single source single sink problem with capacities on the edges 10/39
5 9/39 Major Algorithms - Networks Advogato (Levien) Appleseed (Ziegler and Lausen) MoleTrust (Massa and Avesani) TidalTrust (Golbeck) 10/39 Advogato Levien 2003 Attack resistant Maximum network flow based on Ford-Fulkerson Node capacities determined by the distance from the source This is a single source, multiple sink problem with capacities on the nodes Network flow works on a single source single sink problem with capacities on the edges The graph is transformed