MB-AGCL: multi-behavior adaptive graph contrast learning for recommendation

Abstract Graph Convolutional Networks (GCNs) have achieved remarkable success in recommendation systems by leveraging higher-order neighborhoods. In recent years, multi-behavior recommendation has addressed the challenges of data sparsity and cold start problems to some extent. However, the introduc...

Full description

Saved in:
Bibliographic Details
Main Authors: Xiaowen lv, Yiwei Zhao, Zhihu Zhou, Yifeng Zhang, Yourong Chen
Format: Article
Language:English
Published: Springer 2025-04-01
Series:Complex & Intelligent Systems
Subjects:
Online Access:https://doi.org/10.1007/s40747-025-01880-2
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850272803201220608
author Xiaowen lv
Yiwei Zhao
Zhihu Zhou
Yifeng Zhang
Yourong Chen
author_facet Xiaowen lv
Yiwei Zhao
Zhihu Zhou
Yifeng Zhang
Yourong Chen
author_sort Xiaowen lv
collection DOAJ
description Abstract Graph Convolutional Networks (GCNs) have achieved remarkable success in recommendation systems by leveraging higher-order neighborhoods. In recent years, multi-behavior recommendation has addressed the challenges of data sparsity and cold start problems to some extent. However, the introduction of noise from multi-behavior tasks into the user-item graph exacerbates the impact of noise from a few active users and popularity bias from popular items. To tackle these challenges, graph augmentation has emerged as a promising approach in recommendation systems. However, existing augmentation methods may generate suboptimal graph structures, and maximizing correspondence may capture information unrelated to the recommendation task. To address these issues, we propose a novel approach called the Multi-Behavior Adaptive Graph Contrastive Learning Model (MB-AGCL) for recommendation. Our approach integrates auxiliary behaviors to compensate for data sparsity and utilizes adaptive learning to determine whether to drop edges or nodes, thus obtaining an optimized graph structure that reduces the impact of noise. We then train the original and generated graphs using supervised tasks. Furthermore, we propose an efficient adaptive graph augmentation method that integrates graph augmentation with down-stream tasks to reduce the impact of popularity bias. Finally, we jointly optimize these two tasks. Through extensive experiments on public datasets, we validate the effectiveness of our recommendation model.
format Article
id doaj-art-fa11ea5dacae4ec9839478326500d0ae
institution OA Journals
issn 2199-4536
2198-6053
language English
publishDate 2025-04-01
publisher Springer
record_format Article
series Complex & Intelligent Systems
spelling doaj-art-fa11ea5dacae4ec9839478326500d0ae2025-08-20T01:51:41ZengSpringerComplex & Intelligent Systems2199-45362198-60532025-04-0111611210.1007/s40747-025-01880-2MB-AGCL: multi-behavior adaptive graph contrast learning for recommendationXiaowen lv0Yiwei Zhao1Zhihu Zhou2Yifeng Zhang3Yourong Chen4College of Information Science and Technology, Zhejiang Shuren UniversityCollege of Information Science and Technology, Zhejiang Shuren UniversityCollege of Information Engineering, Zhejiang University of TechnologyTsinghua UniversityCollege of Information Science and Technology, Zhejiang Shuren UniversityAbstract Graph Convolutional Networks (GCNs) have achieved remarkable success in recommendation systems by leveraging higher-order neighborhoods. In recent years, multi-behavior recommendation has addressed the challenges of data sparsity and cold start problems to some extent. However, the introduction of noise from multi-behavior tasks into the user-item graph exacerbates the impact of noise from a few active users and popularity bias from popular items. To tackle these challenges, graph augmentation has emerged as a promising approach in recommendation systems. However, existing augmentation methods may generate suboptimal graph structures, and maximizing correspondence may capture information unrelated to the recommendation task. To address these issues, we propose a novel approach called the Multi-Behavior Adaptive Graph Contrastive Learning Model (MB-AGCL) for recommendation. Our approach integrates auxiliary behaviors to compensate for data sparsity and utilizes adaptive learning to determine whether to drop edges or nodes, thus obtaining an optimized graph structure that reduces the impact of noise. We then train the original and generated graphs using supervised tasks. Furthermore, we propose an efficient adaptive graph augmentation method that integrates graph augmentation with down-stream tasks to reduce the impact of popularity bias. Finally, we jointly optimize these two tasks. Through extensive experiments on public datasets, we validate the effectiveness of our recommendation model.https://doi.org/10.1007/s40747-025-01880-2RecommendationContrastive learningGraph representationSelf-supervised learning
spellingShingle Xiaowen lv
Yiwei Zhao
Zhihu Zhou
Yifeng Zhang
Yourong Chen
MB-AGCL: multi-behavior adaptive graph contrast learning for recommendation
Complex & Intelligent Systems
Recommendation
Contrastive learning
Graph representation
Self-supervised learning
title MB-AGCL: multi-behavior adaptive graph contrast learning for recommendation
title_full MB-AGCL: multi-behavior adaptive graph contrast learning for recommendation
title_fullStr MB-AGCL: multi-behavior adaptive graph contrast learning for recommendation
title_full_unstemmed MB-AGCL: multi-behavior adaptive graph contrast learning for recommendation
title_short MB-AGCL: multi-behavior adaptive graph contrast learning for recommendation
title_sort mb agcl multi behavior adaptive graph contrast learning for recommendation
topic Recommendation
Contrastive learning
Graph representation
Self-supervised learning
url https://doi.org/10.1007/s40747-025-01880-2
work_keys_str_mv AT xiaowenlv mbagclmultibehavioradaptivegraphcontrastlearningforrecommendation
AT yiweizhao mbagclmultibehavioradaptivegraphcontrastlearningforrecommendation
AT zhihuzhou mbagclmultibehavioradaptivegraphcontrastlearningforrecommendation
AT yifengzhang mbagclmultibehavioradaptivegraphcontrastlearningforrecommendation
AT yourongchen mbagclmultibehavioradaptivegraphcontrastlearningforrecommendation