《计算机应用》唯一官方网站 ›› 2022, Vol. 42 ›› Issue (12): 3671-3678.DOI: 10.11772/j.issn.1001-9081.2021101782

• 人工智能 • 上一篇    

融合信任隐含相似度与评分相似度的社会化推荐

周寅莹1, 章梦怡1, 余敦辉1,2, 朱明1()   

  1. 1.湖北大学 计算机与信息工程学院,武汉 430062
    2.湖北省教育信息化工程技术研究中心(湖北大学),武汉 430062
  • 收稿日期:2021-10-18 修回日期:2021-12-19 接受日期:2021-12-23 发布日期:2021-12-31 出版日期:2022-12-10
  • 通讯作者: 朱明
  • 作者简介:周寅莹(1998—),女,湖北随州人,硕士研究生,主要研究方向:知识图谱、推荐系统
    章梦怡(2002—),女,湖北武汉人,主要研究方向:知识图谱
    余敦辉(1974—),男,湖北武汉人,教授,博士,CCF会员,主要研究方向:知识图谱、大数据、服务计算、众包数据管理
  • 基金资助:
    国家重点研发计划项目(2018YFB1003801);国家自然科学基金资助项目(61977021);湖北省技术创新专项(重大项目)(2018ACA13)

Social recommendation combining trust implicit similarity and score similarity

Yinying ZHOU1, Mengyi ZHANG1, Dunhui YU1,2, Ming ZHU1()   

  1. 1.School of Computer and Information Engineering,Hubei University,Wuhan Hubei 430062,China
    2.Hubei Provincial Engineering and Technology Research Center for Education Informationization (Hubei University),Wuhan Hubei 430062,China
  • Received:2021-10-18 Revised:2021-12-19 Accepted:2021-12-23 Online:2021-12-31 Published:2022-12-10
  • Contact: Ming ZHU
  • About author:ZHOU Yinyingborn in 1998, M. S. candidate. Her research interests include knowledge graph, recommendation system.
    ZHANG Mengyi,born in 2002. Her research interests include knowledge graph.
    YU Dunhui,born in 1974, Ph. D., professor. His research interests include knowledge graph, big data, services computing,crowdsourced data management.
  • Supported by:
    National Key Research and Development Program of China(2018YFB1003801);National Natural Science Foundation of China(61977021);Technology Innovation Special Program of Hubei Province (Major Project)(2018ACA13)

摘要:

针对现有的社会化推荐算法大都忽略了物品间的关联关系对推荐精度的影响,并且未能将用户评分与信任数据进行有效结合的问题,提出一种融合信任隐含相似度与评分相似度的社会化推荐算法(SocialTS)。首先,将用户间的评分相似度与信任隐含相似度进行线性组合以得到用户间可靠的相似朋友;然后,将信任关系融入到项目的相关性分析中,从而得到修正后的相似项目;最后,将相似用户、项目作为正则项添加到矩阵分解(MF)模型下,从而获取用户、项目更准确的特征表示。实验结果表明,当潜在特征维度为10时,与主流的社会化推荐算法TrustSVD相比,SocialTS在FilmTrust和CiaoDVD数据集上的均方根误差(RMSE)分别降低了4.23%和8.38%,平均绝对误差(MAE)分别降低了4.66%和6.88%。SocialTS不仅可以有效改善用户冷启动问题,还能较为准确地预测不同评分数量下用户的实际评分,且具有良好的鲁棒性。

关键词: 社会化推荐, 冷启动, 信任隐含相似度, 信任关系, 矩阵分解

Abstract:

Focused on the issue that the most existing social recommendation algorithms ignore the influence of the association relationship between items on the recommendation accuracy, and fail to effectively combine user ratings with trust data, a Social recommendation algorithm combing Trust implicit similarity and Score similarity (SocialTS) was proposed. Firstly, the score similarity and trust implicit similarity between users were combined linearly to obtain reliable similar friends among users. Then, the trust relationship was integrated into the correlation analysis of items, and the modified similar items were obtained. Finally, similar users and items were added to the Matrix Factorization (MF) model as regularization terms, thereby obtaining more accurate feature representations of users and items. Experimental results show that on FilmTrust and CiaoDVD datasets, when the latent feature dimension is 10, compared with the mainstream social recommendation algorithm Trust-based Singular Value Decomposition (TrustSVD), SocialTS has the Root Mean Square Error (RMSE) reduced by 4.23% and 8.38% respectively, and the Mean Absolute Error (MAE) reduced by 4.66% and 6.88% respectively. SocialTS can not only effectively improve users' cold start problem, but also accurately predict users' actual ratings under different numbers of ratings, and has good robustness.

Key words: social recommendation, cold start, trust implicit similarity, trust relationship, Matrix Factorization (MF)

中图分类号: