• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

计算机工程与科学 ›› 2021, Vol. 43 ›› Issue (03): 511-517.

• 人工智能与数据挖掘 • 上一篇    下一篇

基于用户长短期兴趣与知识图卷积网络的推荐

顾军华1,2,佘士耀1,樊帅1,张素琪3   

  1. (1.河北工业大学人工智能与数据科学学院,天津 300401;2.河北省大数据计算重点实验室,天津 300401;

    3天津商业大学信息工程学院,天津 300134)


  • 收稿日期:2020-04-13 修回日期:2020-06-22 接受日期:2021-03-25 出版日期:2021-03-25 发布日期:2021-03-29
  • 基金资助:
    国家自然科学基金(61802282); 天津市企业科技特派员项目(19JCTPJC54200);河北省创新能力提升计划项目(199676146H)

Recommendation based on users’ long- and short-term preference and knowledge graph convolutional network

GU Jun-hua1,2,SHE Shi-yao1,FAN Shuai1,ZHANG Su-qi3   

  1. (1.School of Artificial Intelligence,Hebei University of Technology,Tianjin 300401;

    2.Hebei Province Key Laboratory of Big Data Computing,Tianjin 300401;

    3.School of Information Engineering,Tianjin University of Commerce,Tianjin 300134,China)
  • Received:2020-04-13 Revised:2020-06-22 Accepted:2021-03-25 Online:2021-03-25 Published:2021-03-29

摘要: 基于知识图谱的推荐可以提高推荐的精确性、多样性和可解释性。结合知识图谱与用户长短期兴趣提出了基于用户长短期兴趣与知识图卷积网络的推荐模型(LSKGCN)。在知识图谱推荐算法的基础上提出了将用户长期兴趣偏好与短期兴趣偏好结合的用户表示方法。根据时间筛选近期历史项目并通过知识图卷积网络得到历史项目的向量表示,通过注意力机制得到短期兴趣表示。根据与所有历史项目的最小欧氏距离得到长期兴趣表示。最后在真实数据集MovieLens-20、Amazon Music、 Last.FM上进行测试,验证了该算法的有效性。


关键词: 知识图谱, 推荐系统, 长短期兴趣偏好, 图卷积网络

Abstract: The recommendation based on knowledge graph can improve the accuracy, diversity and interpretability of the recommendation. In this paper, a recommendation medel (LSKGCN) based on the convolution network of knowledge graphs and users’  long- and short-term preference is proposed. Based on the recommendation algorithm of knowledge graph, a user representation method combining long-term preference with short-term preference is proposed. According to the time, the recent history items are screened and the vector representation of the historical items is obtained by the convolution network algorithm of the knowledge graph, and the short-term interest expression is obtained by the attention mechanism. The long-term expression of interest is based on the minimum Euclidean distance from all historical items. Finally, real data sets Movielens-20, Amazon Music, Last.FM are used to test the validity of the algorithm.

Key words: knowledge graph, recommendation system, long- and short-term preference, graph convolutional network