Multi-behavioral recommendation algorithm based on decoupled graph convolution

Xu Yu, Pengju Ding, Jie Yu, Junyu Lin, Lei Guo, Guanfeng Liu, Liang Xi

Abstract: Highlights•Decoupled graph convolution for multi-behavioral recommendation algorithm design.•Divide a multi-behavioral isomorphic graph into multiple non-overlapping ones.•Decoupled graph convolution and contrastive learning to learn feature representation.•Our model outperforms other state-of-the-art methods.

External IDs:dblp:journals/eswa/YuDYLGLX26