Learning multi-behavior user intent for session-based recommendation

Yu Zhang, Xiaoyan Zhu, Guopeng He, Jiaxuan Li, Jiayin Wang

Abstract: Highlights•Multi-behavior global graphs describe each item with different behaviors.•Behavior-aware Global Attention Encoder captures cross-session dependency.•Low-rank self-attention captures in-session local dependency.•User behavior discrimination task solves the user intent entanglement problem.