The Sequence Recommendation Algorithm based on GRU and Attention

Abstract

In recent years, with the booming development of social networks and e-commerce, users have increasingly convenient access to information, while a large amount of data continues to emerge, leading to more and more serious information overload. To alleviate the problem of information overload, recommendation systems have emerged to assist users in sifting through massive amounts of information to find content that meets their needs. Sequential recommendation, as a form of recommendation system, mainly analyzes the interaction behavior between users and items, models user characteristics, and then uses various methods to capture users' long-term and short-term preferences to recommend items of interest to users. Based on the perspective of user preference change over time, this paper provides an in-depth analysis of the current research progress and methods of user behavior sequence recommendation. At the same time, this paper proposes corresponding solution strategies for the problems of cold start, sparse matrix and noise interference faced by traditional recommendation systems. Finally, we will discuss the challenges and future research directions of recommendation systems to provide the theoretical basis for further improvement of recommendation systems.