DAC-GCN: A Dual Actor-Critic Graph Convolutional Network with Multi-Hop Aggregation for Enhanced Recommender Systems

Background: Recommender Systems (RSs) frequently face challenges in balancing exploration and exploitation, particularly in dynamic environments where user behaviors evolve over time. Traditional methods struggle to adapt to these complexities, limiting their effectiveness in real-world domains such...

Full description

Saved in:
Bibliographic Details
Main Authors: Gholamreza Zare, Nima Jafari, Mehdi Hosseinzadeh, Amir Sahafi
Format: Article
Language:English
Published: Prague University of Economics and Business 2025-08-01
Series:Acta Informatica Pragensia
Subjects:
Online Access:https://aip.vse.cz/artkey/aip-202503-0004_dac-gcn-a-dual-actor-critic-graph-convolutional-network-with-multi-hop-aggregation-for-enhanced-recommender-sy.php
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849338715174862848
author Gholamreza Zare
Nima Jafari
Mehdi Hosseinzadeh
Amir Sahafi
author_facet Gholamreza Zare
Nima Jafari
Mehdi Hosseinzadeh
Amir Sahafi
author_sort Gholamreza Zare
collection DOAJ
description Background: Recommender Systems (RSs) frequently face challenges in balancing exploration and exploitation, particularly in dynamic environments where user behaviors evolve over time. Traditional methods struggle to adapt to these complexities, limiting their effectiveness in real-world domains such as e-commerce, streaming services, and social networks. Objective: The objective of this study is to introduce DAC-GCN, a Dual Actor-Critic Graph Convolutional Network, designed to enhance recommendation accuracy, ranking quality, and adaptability to evolving user preferences. DAC-GCN merges graph-based learning with Deep Reinforcement Learning (DRL) techniques to improve both short-term and long-term user-item interactions. Methods: DAC-GCN utilizes a dual architecture with separate Graph Convolutional Networks (GCNs) for policy optimization and value estimation. It incorporates Multi-Hop Aggregation (MHA) to capture extended user-item dependencies and an attention mechanism to emphasize pivotal relationships. We evaluate DAC-GCN on benchmark datasets, including MovieLens 100K, MovieLens 1M, Amazon Subscription Boxes, Amazon Magazine Subscriptions, and Mod Cloth, using standard ranking metrics (Precision@K, Recall@K, NDCG@K, MRR@K, and Hit@K). Results: Experimental results demonstrate that DAC-GCN consistently outperforms state-of-the-art baselines, showing significant improvements in recommendation accuracy, ranking quality, and robustness to shifting user behaviors. The model's ability to capture complex user-item interactions is greatly enhanced by MHA and attention mechanisms, while the dual architecture ensures training stability. Conclusion: DAC-GCN offers a scalable, high-performance solution for modern recommender systems, effectively addressing challenges such as data sparsity and changing user preferences. By integrating graph-based methods with DRL, this study advances both the theory and practice of recommender systems and provides valuable insights for future research and practical applications.
format Article
id doaj-art-cd12db0a20fc4ece8bace0d6e2d42720
institution Kabale University
issn 1805-4951
language English
publishDate 2025-08-01
publisher Prague University of Economics and Business
record_format Article
series Acta Informatica Pragensia
spelling doaj-art-cd12db0a20fc4ece8bace0d6e2d427202025-08-20T03:44:19ZengPrague University of Economics and BusinessActa Informatica Pragensia1805-49512025-08-0114334036410.18267/j.aip.261aip-202503-0004DAC-GCN: A Dual Actor-Critic Graph Convolutional Network with Multi-Hop Aggregation for Enhanced Recommender SystemsGholamreza Zare0https://orcid.org/0000-0002-9184-8538Nima Jafari1https://orcid.org/0000-0002-5514-5536Mehdi Hosseinzadeh2https://orcid.org/0000-0003-1088-4551Amir Sahafi3https://orcid.org/0000-0002-6555-670XDepartment of Computer Engineering, Qeshm Branch, Islamic Azad University, Qeshm, IranDepartment of Computer Engineering, Tabriz Branch, Islamic Azad University, Tabriz, IranPattern Recognition and Machine Learning Laboratory, School of Computing, Gachon University, Seongnam, Republic of KoreaDepartment of Computer Engineering, South Tehran Branch, Islamic Azad University, Tehran, IranBackground: Recommender Systems (RSs) frequently face challenges in balancing exploration and exploitation, particularly in dynamic environments where user behaviors evolve over time. Traditional methods struggle to adapt to these complexities, limiting their effectiveness in real-world domains such as e-commerce, streaming services, and social networks. Objective: The objective of this study is to introduce DAC-GCN, a Dual Actor-Critic Graph Convolutional Network, designed to enhance recommendation accuracy, ranking quality, and adaptability to evolving user preferences. DAC-GCN merges graph-based learning with Deep Reinforcement Learning (DRL) techniques to improve both short-term and long-term user-item interactions. Methods: DAC-GCN utilizes a dual architecture with separate Graph Convolutional Networks (GCNs) for policy optimization and value estimation. It incorporates Multi-Hop Aggregation (MHA) to capture extended user-item dependencies and an attention mechanism to emphasize pivotal relationships. We evaluate DAC-GCN on benchmark datasets, including MovieLens 100K, MovieLens 1M, Amazon Subscription Boxes, Amazon Magazine Subscriptions, and Mod Cloth, using standard ranking metrics (Precision@K, Recall@K, NDCG@K, MRR@K, and Hit@K). Results: Experimental results demonstrate that DAC-GCN consistently outperforms state-of-the-art baselines, showing significant improvements in recommendation accuracy, ranking quality, and robustness to shifting user behaviors. The model's ability to capture complex user-item interactions is greatly enhanced by MHA and attention mechanisms, while the dual architecture ensures training stability. Conclusion: DAC-GCN offers a scalable, high-performance solution for modern recommender systems, effectively addressing challenges such as data sparsity and changing user preferences. By integrating graph-based methods with DRL, this study advances both the theory and practice of recommender systems and provides valuable insights for future research and practical applications.https://aip.vse.cz/artkey/aip-202503-0004_dac-gcn-a-dual-actor-critic-graph-convolutional-network-with-multi-hop-aggregation-for-enhanced-recommender-sy.phprecommender systemgraph convolutional networkactor-criticreinforcement learningmulti-hop aggregationpersonalized recommendations
spellingShingle Gholamreza Zare
Nima Jafari
Mehdi Hosseinzadeh
Amir Sahafi
DAC-GCN: A Dual Actor-Critic Graph Convolutional Network with Multi-Hop Aggregation for Enhanced Recommender Systems
Acta Informatica Pragensia
recommender system
graph convolutional network
actor-critic
reinforcement learning
multi-hop aggregation
personalized recommendations
title DAC-GCN: A Dual Actor-Critic Graph Convolutional Network with Multi-Hop Aggregation for Enhanced Recommender Systems
title_full DAC-GCN: A Dual Actor-Critic Graph Convolutional Network with Multi-Hop Aggregation for Enhanced Recommender Systems
title_fullStr DAC-GCN: A Dual Actor-Critic Graph Convolutional Network with Multi-Hop Aggregation for Enhanced Recommender Systems
title_full_unstemmed DAC-GCN: A Dual Actor-Critic Graph Convolutional Network with Multi-Hop Aggregation for Enhanced Recommender Systems
title_short DAC-GCN: A Dual Actor-Critic Graph Convolutional Network with Multi-Hop Aggregation for Enhanced Recommender Systems
title_sort dac gcn a dual actor critic graph convolutional network with multi hop aggregation for enhanced recommender systems
topic recommender system
graph convolutional network
actor-critic
reinforcement learning
multi-hop aggregation
personalized recommendations
url https://aip.vse.cz/artkey/aip-202503-0004_dac-gcn-a-dual-actor-critic-graph-convolutional-network-with-multi-hop-aggregation-for-enhanced-recommender-sy.php
work_keys_str_mv AT gholamrezazare dacgcnadualactorcriticgraphconvolutionalnetworkwithmultihopaggregationforenhancedrecommendersystems
AT nimajafari dacgcnadualactorcriticgraphconvolutionalnetworkwithmultihopaggregationforenhancedrecommendersystems
AT mehdihosseinzadeh dacgcnadualactorcriticgraphconvolutionalnetworkwithmultihopaggregationforenhancedrecommendersystems
AT amirsahafi dacgcnadualactorcriticgraphconvolutionalnetworkwithmultihopaggregationforenhancedrecommendersystems