Deep Neural Network Technique for Accurate Recommendation Service

With the rapid development of the Internet, we have entered the era of information overload, and it becomes difficult to obtain the information of interest with the explosive growth of data. Fortunately, recommendation algorithm can effectively address this problem in practice. Specifically, the recommendation engines can utilize the existing data to accurately exploit the information that users may be potentially interested in, and then provide personalized recommendation services. In recent years, deep neural networks have been deeply applied in the field of recommendation algorithms. In this paper, we generally review some current recommendation algorithms based on deep neural networks, including the future development direction of recommender systems.