Intra- and Inter-behavior Contrastive Learning for Multi-behavior Recommendation

Abstract

Multi-behavior recommendation (MBR) aims to improve the prediction of target behavior by exploiting multi-typed auxiliary behaviors. However, most MBR suffers from data sparsity in real-world scenarios and thus performs mediocrely. In this paper, we propose a novel Intra- and Inter-behavior Contrastive Learning (IICL) framework to exploit contrastive learning to enhance MBR with two key challenges to be addressed: i). Difficult to learn reliable representations under different behaviors; ii). Sparse supervised signals under target behavior. For the first challenge, we devise an intra-behavior contrastive learning objective, which enhances the representation learning by incorporating potential neighbors into the contrastive learning from the graph topological space and the semantic space, respectively. As for the second challenge, we design an inter-behavior contrastive learning objective, which has the benefit of capturing commonalities between different behaviors and integrating them into the target behavior to alleviate the sparse supervised signal problem. In addition, we also propose an adaptive weight network to customize the integration of all losses efficiently. Extensive experiments on three real-world benchmark datasets show that our proposed method IICL is significantly superior to various state-of-the-art recommendation methods.