A transitivity aware matrix factorization model for recommendation in social networks

摘要: Recommender systems are becoming tools of choice to select the online information relevant a given user. Collaborative filtering is most popular approach building recommender and has been successfully employed in many applications. With advent social networks, network based recommendation emerged. This assumes among users makes recommendations for user on ratings who have direct or indirect relations with As one their major benefits, approaches shown reduce problems cold start users. In this paper, we explore model-based employing matrix factorization techniques. Advancing previous work, incorporate mechanism trust propagation into model principled way. Trust be crucial phenomenon sciences, analysis trust-based recommendation. We conducted experiments two real life data sets. Our demonstrate that modeling leads substantial increase accuracy, particular