Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces Multi-Behavior Latent Factor modeling with Expert network (MBLFE) to address data sparsity in multi-behavior recommendation systems. MBLFE uses a gating expert network to model latent factors, with each expert specializing in a specific factor and the gating network dynamically selecting the optimal expert combination for each user. Self-supervised learning and multi-behavior data enrichment are used to ensure expert independence and factor consistency, leading to significant performance improvements over state-of-the-art baselines on three real-world datasets.
Untangling multiple behavioral factors with an expert network yields more precise user preference representations, significantly boosting recommendation accuracy.
Traditional recommendation methods, which typically focus on modeling a single user behavior (e.g., purchase), often face severe data sparsity issues. Multi-behavior recommendation methods offer a promising solution by leveraging user data from diverse behaviors. However, most existing approaches entangle multiple behavioral factors, learning holistic but imprecise representations that fail to capture specific user intents. To address this issue, we propose a multi-behavior method by modeling latent factors with an expert network (MBLFE). In our approach, we design a gating expert network, where the expert network models all latent factors within the entire recommendation scenario, with each expert specializing in a specific latent factor. The gating network dynamically selects the optimal combination of experts for each user, enabling a more accurate representation of user preferences. To ensure independence among experts and factor consistency of a particular expert, we incorporate self-supervised learning during the training process. Furthermore, we enrich embeddings with multi-behavior data to provide the expert network with more comprehensive collaborative information for factor extraction. Extensive experiments on three real-world datasets demonstrate that our method significantly outperforms state-of-the-art baselines, validating its effectiveness.