Using Social Tag Embedding in a Collaborative Filtering Approach for Recommender Systems

Nowadays, the use of social information is extending to more and more application domains. In the field of recommender systems, this information has been exploited in different ways to address some problems, especially associated with collaborative filtering methods, and thus achieve more reliable recommendations. Specifically, social tagging is used in this area mainly to characterize the items that are the subject of the recommendations. In this work, a user-based collaborative filtering approach is presented, where tags processed by word embedding techniques are used to characterize users. User similarities based on both tag embedding and ratings are combined to generate the recommendations. In the study conducted on two popular datasets, the reliability of this approach for rating prediction and top-N recommendations was tested, showing the best performance against the most widely used collaborative filtering methods.