Exploring global information for session-based recommendation

Ziyang Wang, Wei Wei, Ding Zou, Yifan Liu, Xiao-Li Li, Xian-Ling Mao, Minghui Qiu

Abstract: Highlights•This paper explores high-order global information for session-based recommendation.•We leverage prototype contrastive learning to learn the global item transition.•We compare two versions of the global exploration mode in detail.•We conduct extensive experiments to compare the performance of each model.