Computer Engineering and Applications ›› 2023, Vol. 59 ›› Issue (15): 264-273.DOI: 10.3778/j.issn.1002-8331.2209-0083

• Big Data and Cloud Computing • Previous Articles     Next Articles

Session-Based Recommendation Based on Multi-Graph Neural Network Incorporating Social Information

LEI Jingsheng, LI Ran, YANG Shengying, SHI Wenbin   

  1. School of Information and Electronic Engineering, Zhejiang University of Science and Technology, Hangzhou 310023, China
  • Online:2023-08-01 Published:2023-08-01

融合社交信息的多图神经网络会话推荐方法

雷景生,李冉,杨胜英,史文彬   

  1. 浙江科技学院 信息与电子工程学院,杭州 310023

Abstract: In recommender systems, users’ interest in items is dynamic and influenced by various factors such as their own historical behavior and friends’ behavior. It has been a challenge for recommendation algorithms to jointly model users’ dynamic interests and social relationships. In this paper, the dynamic interests of users are explored by partitioning their behaviors into session sequences and modeling them as global graphs. After that, a social relationship graph based on users’ social relationships is constructed, and then the influence of users’ social relationships is captured through graph attention networks to dynamically determine the influence of each friend, and users’ dynamic interests are combined with friends’ social influence to obtain the final recommendation results. The algorithm is validated on Douban, Delicious and Yelp datasets. Compared with the most available baseline model, the algorithm improves more than 6 percentage points on the Douban dataset and more than 3 percentage points on the Delicious and Yelp datasets, which proves the effectiveness of the algorithm.

Key words: session-based recommendation, graph neural network, attentional mechanisms, social information, personalized preferences

摘要: 在推荐系统中,用户对于项目的兴趣是动态变化的且会受到自身历史行为以及朋友行为等多种因素的影响。而如何对用户的动态兴趣以及社交关系进行共同建模一直是推荐算法的一个挑战。通过将用户的行为分割为会话序列,并建模为全局图来探索用户的动态兴趣。之后根据用户的社交关系构建社交关系图,再通过图注意力网络捕获用户社交关系的影响,动态确定每个朋友的影响力,并将用户的动态兴趣与朋友的社交影响结合以得到最终的推荐结果。算法在Douban、Delicious和Yelp数据集上进行了验证,相较最优的基线模型,算法在Douban数据集各项指标上提高超过6个百分点,在Delicious和Yelp数据集各项指标上提高超过3个百分点,证明了算法的有效性。

关键词: 会话推荐, 图神经网络, 注意力机制, 社交信息, 个性化偏好