计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (6A): 230800160-7.doi: 10.11896/jsjkx.230800160

• 大数据&数据科学 • 上一篇    下一篇

RM-RT2NI:融合评论时效与可信近邻影响力的推荐模型

韩志耕1,2, 周婷1,2, 陈耿2,3, 付纯硕1,2, 陈健1,2   

  1. 1 南京审计大学计算机学院/智能审计学院 南京 211815
    2 江苏省审计信息工程与技术协同创新中心 南京 211815
    3 南京审计大学会计学院 南京 211815
  • 发布日期:2024-06-06
  • 通讯作者: 周婷(1442677601@qq.com)
  • 作者简介:(hanzgnit@126.com)
  • 基金资助:
    国家自然科学基金(72072091);江苏省高校自然科学研究基金(21KJA520002,22KJA520005);江苏省研究生科研与实践创新计划项目(KYCX23_2345);江苏省审计信息工程与技术协同创新中心项目

RM-RT2NI:A Recommendation Model with Review Timeliness and Trusted Neighbor Influence

HAN Zhigeng1,2, ZHOU Ting1,2, CHEN Geng2,3, FU Chunshuo1,2, CHEN Jian1,2   

  1. 1 School of Computer Science/School of Intelligence Audit,Nanjing Audit University,Nanjing 211815,China
    2 Jiangsu Provincial Couaborative Innovation Center for Audit Information Engineering and Technology,Nanjing 211815,China
    3 School of Accounting,Nanjing Audit University,Nanjing 211815,China
  • Published:2024-06-06
  • About author:HAN Zhigeng,born in 1976,Ph.D,associate professor,is a member of CCF(No.20278M).His main research interests include trustworthy recommendations and intelligent audit.
    ZHOU Ting,born in 1999,postgra-duate.Her main research interests include trustworthy recommendations and intelligent audit.
  • Supported by:
    National Natural Science Foundation of China(72072091),Natural Science Foundation of the Higher Education Institutions of Jiangsu Province,China(21KJA520002,22KJA520005),Postgraduate Research & Practice Innovation Program of Jiangsu Province(KYCX23_2345) and Projects of Jiangsu Provincial Collaborative Innovation Center for Audit Information Engineering and Technology.

PDF (PC)

268

摘要： 基于矩阵分解的推荐模型虽然能够处理高维评分数据,但容易遭受评分数据稀疏性的困扰。基于评分和评论的推荐模型通过外加隐藏在评论中的用户偏好与物品属性信息,缓解了评分数据的稀疏性,但在特征提取时大多没有关注评论时效性和可信近邻影响力,无法获得更丰富的用户和物品特征。为进一步提高推荐精度,提出了融合评论时效与可信近邻影响力的推荐模型RM-RT2NI。基于评分矩阵,该模型使用矩阵分解提取了用户偏好和物品属性的浅层特征,利用云模型和修正的用户相似度评估模型和新构建的信度评估模型提取出可信近邻影响力;基于评论文本,该模型利用BERT模型获得每条评论的隐表达,利用双向GRU提取评论间的联系,利用新构建的融合时间因子的注意力机制识别各评论的时效贡献度,以获取用户和物品的深层特征。在此基础上,将用户浅层特征、深层特征以及可信近邻影响力特征融合成用户特征,将物品浅层特征和深层特征融合成物品特征,并将它们输入全连接神经网络以预测用户-物品评分。在5组公开数据集上对RM-RT2NI的推荐性能进行了实验评估,结果显示,与7个基线模型相比,RM-RT2NI具有更高的评分预测精度,且RMSE平均降低了3.0657%。

关键词: 推荐模型, 评分矩阵, 评论文本, 评论时效, 可信近邻影响力, 多特征融合

Abstract: While recommendation models based on matrix factorization can handle high-dimensional rating data,they are prone to challenges posed by data sparsity in ratings.Recommendation models that incorporate both ratings and reviews alleviate the sparsity issue by incorporating latent user preferences and item attribute information embedded in reviews.However,these models often neglect the review timeliness and the trusted neighbor influence during feature extraction,resulting in limited acquisition of comprehensive user and item characteristics.In order to enhance accuracy further,a novel recommendation model named RM-RT2NI is proposed,which integrates the review timeliness and the trusted neighbor influence.Built upon the rating matrix,this model employs matrix factorization to extract shallow features representing user preferences and item attributes.It employs cloud mo-deling,a refined user similarity assessment model,and a newly constructed credibility assessment model to capture the trusted neighbor influence.Leveraging the textual content of reviews,BERT is utilized to obtain latent representations of individual reviews.Bi-directional GRU is employed to capture inter-review relationships,while an attention mechanism incorporating timeliness is introduced to evaluate the timeliness contribution of each review,thus deriving deep features for users and items.Subsequently,the shallow and deep user features,along with the credibility-enhanced neighboring influence features,are fused to form comprehensive user representations.Similarly,shallow and deep item features are merged with this fused representation to gene-rate comprehensive item representations.These representations are then fed into a fully connected neural network to predict user-item ratings.Experimental evaluation is conducted on five publicly available datasets.The results demonstrate that,in comparison to seven baseline models,RM-RT2NI exhibits superior rating prediction accuracy,yielding an average RMSE reduction of 3.0657%.

Key words: Recommendation model, Rating matrix, Review text, Review timeliness, Trusted neighbor influence, Multi-feature fusion

中图分类号: 

  • TP391
[1]KO H,LEE S,PARK Y,et al.A survey of recommendation systems:recommendation models,techniques,and application fields[J].Electronics,2022,11(1):141-189.
[2]ZHU Z,WANG S,WANG F,et al.Recommendation networks of homogeneous products on an E-commerce platform:Measurement and competition effects[J].Expert Systems with Applications,2022,201:117-128.
[3]DHELIM S,AUNG N,BOURASM A,et al.A survey on personality-aware recommendation systems[J].Artificial Intelligence Review,2022,55:2409-2454.
[4]RAZA S,DING C.News recommender system:a review of recent progress,challenges,and opportunities[J].Artificial Intelligence Review,2022,55:749-800.
[5]ALHIJAWI B,AWAJAN A,FRAIHAT S.Survey on the objectives of recommender systems:Measures,solutions,evaluation methodology,and new perspectives[J].ACM Computing Surveys,2022,55(5):1-38.
[6]TEGENE A,LIU Q,GAN Y,et al.Deep Learning and Embedding Based Latent Factor Model for Collaborative Recommender Systems[J].Applied Sciences,2023,13(2):726.
[7]ZHENG L,NOROOZI V,YU P S.Joint deep modeling of users and items using reviews for recommendation[C]//Proceedings of the tenth ACM International Conference on Web Search and Data Mining.New York,USA,2017:425-434.
[8]CHEN C,ZHANG M,LIU Y,et al.Neural attentional rating regression with review-level explanations[C]//Proceedings of the 2018 World Wide Web Conference.Republic and Canton of Geneva,CHE,2018:1583-1592.
[9]FENG X J,ZENG Y Z.Joint Deep Modeling of Rating Matrix and Reviews for Recommendation[J].Chinese Journal of Computers,2020,43(5):884-900.
[10]LI S Z,YU L T,DENG X H.Recommendation Model Combining Deep Sentiment Analysis and Scoring Matrix[J].Journal of Electronics & Information Technology,2022,44(1):245-253.
[11]ZHAI Y,ZHANG X,DAO R,et al.Research on the usefulness of online reviews in catering trade[C]//2017 3rd International Conference on Information Management(ICIM).Chengdu,China,2017:247-251.
[12]YUAN Y,ZAHIR A,YANG J.Modeling implicit trust in matrix factorization-based collaborative filtering[J].Applied Sciences,2019,9(20):4378-4396.
[13]KASHANI S M Z,HAMIDZADEH J.Improvement of non-ne-gative matrix-factorization-based and Trust-based approach to collaborative filtering for recommender systems[C]//2020 6th Iranian Conference on Signal Processing and Intelligent Systems(ICSPIS).Mashhad,Iran,2020:1-7.
[14]XU S,ZHUANG H,SUN F,et al.Recommendation algorithm of probabilistic matrix factorization based on directed trust[J].Computers & Electrical Engineering,2021,93:107-206.
[15]XI W D,HUANG L,WANG C D,et al.Deep rating and review neural network for item recommendation[J].IEEE Transactions on Neural Networks and Learning Systems,2021,33(11):6726-6736.
[16]WU L,HE X,WANG X,et al.A survey on accuracy-oriented neural recommendation:From collaborative filtering to information-rich recommendation[J].IEEE Transactions on Knowledge and Data Engineering,2023,35(5):4425-4445.
[17]FANG H,ZHANG D,SHU Y,et al.Deep learning for sequential recommendation:Algorithms,influential factors,and evaluations[J].ACM Transactions on Information Systems(TOIS),2020,39(1):1-42.
[18]TANG H,LEI M,GONG Q,et al.A BP neural network recommendation algorithm based on cloud model[J].IEEE Access,2019,7:35898-35907.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发