Deep learning with consumer preferences for recommender system

With the arrival of big data era and the fast development of E-commerce, recommender systems(RSs) are used in more and more application domains to assist customers in the search for their favorite products. Collaborative filtering(CF) is one of the most successful and widely used recommendation approaches. Poor recommender quality is a major challenge in traditional CF. And one reason causing this circumstance is the sparsity of data. In this paper, we propose to a deep learning model to predict the values of null ratings. Moreover, we investigate personal recommendation based on customer preferences and search the neighbors through the customer preferences. Accordingly, it has effectively solved recommendation quality problems of traditional CF algorithm under the condition of data sparse and poor result quality.