Sequential Equipment Recommendation Algorithm based on Context Awareness

• 1、Beijing Key Laboratory of Intelligent Telecommunications Software and Multimedia(Beijing University of Posts and Telecommunications), Beijing 100876

Abstract：In recent years, recommendation systems have been used to improve user experience in video games. In Multiplayer Online Battle Arena games, recommendation systems help the user choose the right equipment for the hero during a match. The current equipment recommendation approach focuses on recommending a fixed set of equipment. However, matchups in MOBA games are a sequential process, and the hero\'s past decisions affect the current ones. Therefore, this paper proposes a sequential MOBA equipment recommendation algorithm based on contextual information. It consists of a context encoder and a sequence encoder. The encoded representation of heroes and equipment is generated by the context encoder, which is related to the context information in the peer. A sequential encoder captures sequential patterns of data to recommend the next device. Experimental results on real data sets show that the proposed method improves the recommendation performance.

Keywords： computer application technology sequential recommendation context information equipment recommendation