《计算机应用》唯一官方网站 ›› 2023, Vol. 43 ›› Issue (4): 1043-1049.DOI: 10.11772/j.issn.1001-9081.2022040481

• 人工智能 • 上一篇    

基于改进胶囊网络的会话型推荐模型

孙浩1,2(), 曹健1,2, 李海生1,2, 毛典辉1,2   

  1. 1.北京工商大学 计算机学院,北京 100048
    2.食品安全大数据技术北京市重点实验室(北京工商大学),北京 100048
  • 收稿日期:2022-04-14 修回日期:2022-08-02 接受日期:2022-08-03 发布日期:2022-09-27 出版日期:2023-04-10
  • 通讯作者: 孙浩
  • 作者简介:曹健(1982—),男,山东临沂人,副教授,博士,主要研究方向:机器学习、图像处理;
    李海生(1974-),男,山东德州人,教授,博士生导师,博士,CCF会员,主要研究方向:机器学习、图形学;
    毛典辉(1979-),男,湖北浠水人,教授,博士生导师,博士,主要研究方向:人工智能、区块链。
  • 基金资助:
    国家自然科学基金资助项目(61877002);北京市教委-市自然基金委联合资助项目(KZ202110011017);北京市自然科学基金-丰台轨道交通前沿研究联合基金资助项目(L191009)

Session-based recommendation model based on enhanced capsule network

Hao SUN1,2(), Jian CAO1,2, Haisheng LI1,2, Dianhui MAO1,2   

  1. 1.School of Computer Science and Engineering,Beijing Technology and Business University,Beijing 100048,China
    2.Beijing Key Laboratory of Big Data Technology for Food Safety (Beijing Technology and Business University),Beijing 100048,China
  • Received:2022-04-14 Revised:2022-08-02 Accepted:2022-08-03 Online:2022-09-27 Published:2023-04-10
  • Contact: Hao SUN
  • About author:CAO Jian, born in 1982, Ph. D., associate professor. His research interests include machine learning, image processing.
    LI Haisheng, born in 1974, Ph. D., professor. His research interests include machine learning, graphics.
    MAO Dianhui, born in 1979, Ph. D., professor. His research interests include artificial intelligence, blockchain.
  • Supported by:
    National Natural Science Foundation of China(61877002);Program Jointly Funded by Beijing Municipal Education Commission and Beijing Natural Science Foundation(KZ202110011017);Beijing Natural Science Foundation and Fengtai Rail Transit Frontier Research Fund(L191009)

摘要:

针对现有的会话型推荐模型难以从简短的会话中捕获项目之间的依赖关系的问题,在考虑了复杂的项目交互和动态的用户兴趣变化后,提出了一种基于会话型推荐的改进胶囊网络(SR-ECN)模型。首先,利用图神经网络(GNN)处理会话序列数据,以得到每个项目嵌入向量;然后,利用胶囊网络的动态路由机制,从交互历史中聚合高级用户的偏好;此外,所提模型引入自注意力网络进一步考虑用户和项目的潜在信息,从而为用户推荐更合适的项目。实验结果表明,在Yoochoose数据集上,所提模型的召回率和平均倒数排名(MRR)均优于SR-GNN(Session-based Recommendation with GNN)、TAGNN(Target Attentive GNN)等所有对比模型,与基于无损边缘保留聚合和快捷图注意力的推荐(LESSR)模型相比,所提模型的召回率和MRR分别提升了0.92和0.45个百分点,验证了改进胶囊网络对用户兴趣偏好提取的有效性。

关键词: 胶囊网络, 会话型推荐, 图神经网络, 自注意力机制, 推荐系统

Abstract:

Aiming at the dependencies between items are difficult to be captured by the present session-based recommendation models from short sessions, with complex item interactions and dynamic user interest changes considered, a Session-based Recommendation of Enhanced Capsule Network (SR-ECN) model was proposed. First, session sequence data was processed by using the Graph Neural Network (GNN) to obtain embedded vector of each item. Then, the dynamic routing mechanism of the capsule network was used to aggregate high-level user preferences from the interaction history. In addition, a self-attention network was introduced by the proposed model to further consider potential information about users and items, thereby recommending more suitable items for users. Experimental results show that, on Yoochoose dataset, the proposed model is superior to all comparison models such as Session-based Recommendation with GNN (SR-GNN), Target Attentive GNN (TAGNN), and the proposed model improves 0.92 and 0.45 percentage points compared to the Lossless Edge-order preserving aggregation and Shortcut graph attention for Session-based Recommendation (LESSR) model in terms of recommendation recall and Mean Reciprocal Rank (MRR) respectively.

Key words: capsule network, session-based recommendation, Graph Neural Network (GNN), self-attention mechanism, recommender system

中图分类号: