CogTime_RMF: regularized matrix factorization with drifting cognition degree for collaborative filtering

摘要: Due to the exponential growth of information, recommender systems have been a widely exploited technique solve problem information overload effectively. Collaborative filtering (CF) is most successful and extensively employed recommendation approach. However, current CF methods recommend suitable items for users mainly by user-item matrix that contains individual preference in collection. So these suffer from such problems as sparsity available data low accuracy predictions. To address issues, borrowing idea cognition degree cognitive psychology employing regularized factorization (RMF) basic model, we propose novel drifting degree-based RMF collaborative method named CogTime_RMF incorporates both users' with time. Moreover, conduct experiments on real datasets MovieLens 1 M 100 k, compared three similarity based other latest methods. Empirical results demonstrate our proposal can yield better performance over recommendation. In addition, show alleviate sparsity, particularly circumstance few ratings are observed.