《计算机应用》唯一官方网站 ›› 2021, Vol. 41 ›› Issue (11): 3164-3170.DOI: 10.11772/j.issn.1001-9081.2021010060

• 人工智能 • 上一篇    下一篇

基于会话的多粒度图神经网络推荐模型

任俊伟1, 曾诚1,2,3(), 肖丝雨1, 乔金霞1, 何鹏1,2,3   

  1. 1.湖北大学 计算机与信息工程学院,武汉 430062
    2.湖北省软件工程工程技术研究中心(湖北大学),武汉 430062
    3.智慧政务与人工智能应用湖北省工程研究中心(湖北大学),武汉 430062
  • 收稿日期:2021-01-13 修回日期:2021-04-23 接受日期:2021-04-29 发布日期:2021-05-12 出版日期:2021-11-10
  • 通讯作者: 曾诚
  • 作者简介:任俊伟(1992—),男,湖北宜昌人,硕士研究生,主要研究方向:知识图谱、推荐系统
    曾诚(1976—),男,湖北武汉人,教授,博 士,CCF会员,主要研究方向:人工智能、计算机软件
    肖丝雨(1994—),女,湖北武汉人,硕士研究生,主要研究方向:知识图谱、推荐系统
    乔金霞(1997—),女,山西高平人,硕士研究生,主要研究方向:知识图谱、推荐系统
    何鹏(1988—),男,湖北武汉人,副教授,博士,主要研究方 向:大数据处理、软件度量、复杂网络。
  • 基金资助:
    国家自然科学基金面上项目(61977021)

Session-based recommendation model of multi-granular graph neural network

Junwei REN1, Cheng ZENG1,2,3(), Siyu XIAO1, Jinxia QIAO1, Peng HE1,2,3   

  1. 1.School of Computer Science and Information Engineering,Hubei University,Wuhan Hubei 430062,China
    2.Engineering and Technical Research Center of Hubei Province in Software Engineering (Hubei University),Wuhan Hubei 430062,China
    3.Engineering Research Center of Hubei Province in Government Affairs and Application of Artificial Intelligence (Hubei University),Wuhan Hubei 430062,China
  • Received:2021-01-13 Revised:2021-04-23 Accepted:2021-04-29 Online:2021-05-12 Published:2021-11-10
  • Contact: Cheng ZENG
  • About author:REN Junwei,born in 1992,M. S. candidate. His research interests include knowledge graph,recommender system
    ZENG Cheng,born in 1976,Ph. D.,professor. His research interests include artificial intelligence,computer software
    XIAO Siyu,born in 1994,M. S. candidate. Her research interests include knowledge graph,recommender system
    QIAO Jinxia, born in 1997, M. S. candidate. Her research interests include knowledge graph,recommender system
    HE Peng,born in 1988,Ph. D.,associate professor. His research interests include big data processing,software metrics,complex network.
  • Supported by:
    the Surface Program of National Natural Science Foundation of China(61977021)

摘要:

基于会话的推荐旨在根据当前用户的匿名会话的点击序列信息来预测用户的下一次点击行为。现有方法多数都是通过对用户会话点击序列的物品信息进行建模,并学习物品的向量表示,进而进行推荐。而作为一种粗粒度的信息,物品的类别信息对物品有聚合作用,可作为物品信息的重要补充。基于此,提出了基于会话的多粒度图神经网络推荐模型(SRMGNN)。首先,使用图神经网络(GNN)得到会话序列中的物品和物品类别的嵌入向量表示,并使用注意力网络捕捉用户的注意力信息;然后,将赋予了不同注意力权重值的物品和物品类别信息进行融合后,输入到门限循环单元(GRU)里;最后,通过GRU学习会话序列的物品时序信息,并给出推荐列表。在公开的Yoochoose数据集和Diginetica数据集上进行实验,实验结果验证了该模型在增加了物品类别信息后的优势,且实验结果表明了在Precision@20和MRR@20这2种评价指标上,该模型相较于短期注意力/记忆优先级(STAMP)模型、神经注意力(NARM)模型、GRU4REC等8种模型均有更好的效果。

关键词: 基于会话的推荐, 多粒度, 推荐模型, 图神经网络, 点击序列

Abstract:

Session-based recommendation aims to predict the user’s next click behavior based on the click sequence information of the current user’s anonymous session. Most of the existing methods realize recommendations by modeling the item information of the user’s session click sequence and learning the vector representation of the items. As a kind of coarse-grained information, the item category information can aggregate the items and can be used as an important supplement to the item information. Based on this, a Session-based Recommendation model of Multi-granular Graph Neural Network (SRMGNN) was proposed. Firstly, the embedded vector representations of items and item categories in the session sequence were obtained by using the Graph Neural Network (GNN), and the attention information of users was captured by using the attention network. Then, the items and item category information given by different weight values of attention were fused and input into the Gated Recurrent Unit (GRU). Finally, through GRU, the item time sequence information of the session sequence was learned, and the recommendation list was given. Experiments performed on the public Yoochoose dataset and Diginetica dataset verify the advantages of the proposed model with the addition of item category information, and show that the model has better effect compared with all the eight models such as Short-Term Attention/Memory Priority (STAMP), Neural Attentive session-based RecomMendation (NARM), GRU4REC on the evaluation indices Precision@20 and Mean Reciprocal Rank (MRR)@20.

Key words: session-based recommendation, multi-granular, recommendation model, Graph Neural Network (GNN), click sequence

中图分类号: