Tag and Resource-Aware Collaborative Filtering Algorithms for Resource Recommendation

Recommender systems suggest resources to users based on collaborative filtering techniques, typically by exploiting correlations between individual user ratings of the resources they are interested in. Tags are a new form of metadata increasingly used in social bookmarking sites by users to annotate bookmarked resources. Our goal is to harness the implicit knowledge contained in these tags to improve the quality of recommendations to users. We use both tag and resource-interest knowledge in our user-based collaborative filtering algorithms to profile users and compute similarity between them. Sparsity is a challenge which occurs in a Social Recommendation System when the number of tags and resources to profile a user are inadequate to provide good quality recommendations. To address this problem, we designed a Tripartite Nearest Neighbor Algorithm (TRNNA) which combines three views of the data: the tags (TNNA), the resources (RNNA) and the collection of tags for a resource (Resource Vector of Tags or RVTA). TRNNA computes distance between users based on Cosine Similarity, which in turn is used to provide a high quality recommendation of resources. Our empirical evaluation, based on a user study in which research papers were recommended to participants and relevance of recommendation was evaluated, indicates that TRNNA and RNNA provide better recommendation than TNNA and RVTA.