Abstract:Fusion of multi-source information is one of the primary methods to alleviate data sparsity in recommender systems (RS). Hypergraphs have shown remarkable capabilities in dealing with the diversity of multi-source information, especially in modeling high-order user-item relationships. However, most hypergraph research focuses on constructing hypergraphs using a particular type of hyperedge, which might not fully capture the high-order implicit associations among heterogeneous information sources. Furthermore, existing Hypergraph Attention Networks (Hyper-GAT), mainly emphasizes on information propagation between nodes and hyperedges, insufficiently exploring density information within complex hyperedges. Moreover, when hypergraphs combine data from multiple heterogeneous graphs, redundant information from different viewpoints can reduce the effectiveness of multivariate modeling hyperedges. Thus, we propose a recommendation algorithm based on multi-source heterogeneous hypergraphs and contrastive learning (MHCLR), which improves recommendation accuracy through multi-source information fusion and higher-order information correlation. First, multi-source heterogeneous hypergraphs (MHC) are generated by combining distance, behavior, attribute, and prediction hyperedges. We mine associations among multi-source information, enhancing high-order semantic connections among users and items. Then, a Spatial Density Hypergraph Attention Network (SD-HGAT) is proposed based on composite hyperedges, which enriches the user and item embedding representations by focusing on nodes and hyperedges density. Finally, we design a multiple cross view contrastive learning (MCC) that compares views centered around knowledge graphs with hypergraphs, improving the accuracy of multivariate relationship modeling and enabling multi-level user profiling construction. It is observed that MHCLR outperforms the baselines in terms of Recall, Precision, and NDCG based on the experimental results by Yelp, Last-FM, and Douban.

HIN-based Rating Prediction in Recommender Systems Via GCN and Meta-Learning

Attention-aware Multi-channel Graph Convolutional Rating Prediction Model for Heterogeneous Information Networks

A Tri-Attention Neural Network Model-BasedRecommendation.

All-in-One: Heterogeneous Interaction Modeling for Cold-Start Rating Prediction

Heterogeneous Neural Attentive Factorization Machine for Rating Prediction.

Heterogeneous Information Network Embedding for Recommendation

Multi-Task Learning for Recommendation Over Heterogeneous Information Network

Meta-path Embedding Based Recommendation over Heterogeneous Information Network

Contrastive Meta-learning on Heterogeneous Information Networks for Cold-start Recommendation

Automatic Meta-Path Discovery for Effective Graph-Based Recommendation

MHRR: MOOCs Recommender Service With Meta Hierarchical Reinforced Ranking

An Effective Two-way Metapath Encoder over Heterogeneous Information Network for Recommendation

Meta-Graph Based Attention-Aware Recommendation Over Heterogeneous Information Networks

Meta-Graph Based Recommendation Fusion over Heterogeneous Information Networks.

Boosting Meta-Learning Cold-Start Recommendation with Graph Neural Network

Combining Meta-Graph and Attention for Recommendation over Heterogenous Information Network

Meta-relation assisted knowledge-aware coupled graph neural network for recommendation

A Recommendation Method Based on Multi-Source Heterogeneous Hypergraphs and Contrastive Learning

Dynamic Offset Metric on Heterogeneous Information Networks for Cold-start Recommendation.

Relation-propagation Meta-Learning on an Explicit Preference Graph for Cold-Start Recommendation

Meta-Path Based Neighbors for Behavioral Target Generalization in Sequential Recommendation