Combining Positive and Negative Feedbacks with Factored Similarity Matrix for Recommender Systems

Traditional collaborative filtering algorithms like ItemKNN, cannot capture the relationships between items that are not co-rated by at least one user. To cope with this problem, the item-based factor models are put forward to utilize low dimensional space to learn implicit relationships between items. However, these models consider all user’s rated items equally as positive examples, which is unreasonable and fails to interpret the actual preferences of users. To tackle the aforementioned problems, in this paper, we propose a novel item-based latent factor model, which can consider user’s positive and negative feedbacks while learning item-item correlations. In particular, for each user, we divide his rated items into two different parts, i.e., positive examples and negative examples, depending on whether the rating of the item is above the average rating of the user or not. In our model, we assume that the predicted rating of an item should be boosted if the item is similar to most of the positive examples. On the contrary, the predicted rating should be diminished if the item is similar to most of the negative examples. The item-item similarity is approximated by an inner product of two low-dimensional item latent factor matrices which are learned using a structural equation modeling approach. Comprehensive experiments on two benchmark datasets indicate that our method has significant improvements as compared with existing approaches in both rating prediction and top-N recommendation.