A Survey on Sequential Recommendation

In modern e-commerce, the user's sequential behavior contains a wealth of information. According to the user's historical interaction behavior information, modeling the user's preferences to predict the continuous items that the user may interact with is the main task of the sequential recommendation system. In this article, we conducted a systematic and comprehensive review of sequential recommendation systems based on user interaction behavior. Specifically, first of all, we briefly introduce the existing general recommendation methods, and point out the shortcomings of the general recommendation method, and then systematically expound the sequence recommendation, and finally summarize and forecast.