A Survey and Critique of Deep Learning on Recommender Systems

Recommender systems have become extremely common in recent years. Companies, such as Amazon or eBay, developed a large number products to meet different needs of customers. A increasing number of options are available to customers in the era of E-commerce. Thus, in this new level of customization, in order to find what they really need, customers must process a large amount of information provided by businesses. One solution to ease this overload problem is recommender systems. On one hand, traditional recommender systems recommend items based on different criteria, such as the past preference of users or user profiles. On the another hand, deep learning techniques achieve promising performance in various areas, such as Computer Vision, Audio Recognition and Natural Language Processing. However, applications of deep learning in recommender systems have not been well explored yet. In this article, we firstly introduce traditional techniques involved in recommender systems and deep learning in the first chapter. And then, a survey and critique of several state-of-the-art deep recommendation systems will be provided in the following chapters.