计算机科学

计算机科学 ›› 2023, Vol. 50 ›› Issue (11A): 220800205-8.doi: 10.11896/jsjkx.220800205

• 大数据&数据科学 • 上一篇    下一篇

基于全局特征增强的会话推荐算法

靳博文1,2, 王庆梅1,2, 胡承佐1, 魏嘉呈1   

  1. 1 北京科技大学国家材料服役安全科学中心 北京 100083
    2 南方海洋科学与工程广东省实验室(珠海)海洋工程材料与腐蚀控制创新团队 广东 珠海 519080
  • 发布日期:2023-11-09
  • 通讯作者: 王庆梅(qmwang@ustb.edu.cn)
  • 作者简介:(m202121149@xs.ustb.edu.cn)
  • 基金资助:
    南方海洋科学与工程广东省实验室(珠海)创新团队建设项目(311020012)

Global Feature Enhanced for Session-based Recommendation

JIN Bowen1,2, WANG Qingmei1,2, HU Chengzuo1, WEI Jiacheng1   

  1. 1 National Center for Materials Service Safety,University of Science and Technology Beijing,Beijing 100083,China
    2 Innovation Group of Marine Engineering Materials and Corrosion Control,Southern Marine Science and Engineering Guangdong Laboratory,Zhuhai,Guangdong 519080,China
  • Published:2023-11-09
  • About author:JIN Bowen,born in 1999,postgraduate.His main research interests include data mining and recommendation.
    WANG Qingmei,born in 1975,Ph.D,associate researcher.Her main research interests include deep learning and computer vision.
  • Supported by:
    Innovation Group Project of Southern Marine Science and Engineering Guangdong Laboratory(Zhuhai)(311020012).

PDF (PC)

538

摘要： 基于会话的推荐系统的研究通常侧重于在使用浅层神经网络的同时通过聚集节点的K跳邻域来对当前会话用户偏好建模,但是此类方法面临着过平滑的问题。为此,提出了一种基于全局特征增强的会话推荐算法(GFE-SR)。首先,该方法在会话图中利用图神经网络和注意力机制获得会话级项目表示。其次,在全局图的特征传播阶段给每个节点的最近邻域按比例赋予权重来限制过平滑问题,通过全局图进行特征表征融合,获得特征增强的全局级项目表示。然后,通过注意力机制聚合两种项目表示对当前会话的用户偏好进行建模,最终输出候选项目的预测概率。在3个基准数据集上的实验表明,该方法的性能优于现有的最佳方法如 GCE-GNN 等,最高可提升5.2%,证明了该方法的有效性。

关键词: 会话推荐, 推荐系统, 图卷积网络, 注意力机制, 过平滑

Abstract: Most session-based recommendation system researches commonly model current session user preferences by aggregating K-hop neighborhoods of nodes while using shallow neural networks.However,such methods face the problem of over-smoo-thing.This paper proposes global feature enhanced for session-based recommendation(GFE-SR).Firstly,this method utilizes graph neural network and attention mechanism to obtain session-level item representations.Then,in feature propagation stage of the global graph,the nearest neighbors of each node are proportionally weighted to limit over-smoothing to obtain the feature-enhanced global-level item representations.These two item representations are aggregated through an attention mechanism to model current session user preferences.The final output is the probability of the candidate item.Experiments on three public datasets show that this method outperforms the state-of-the-art methods such as GCE-GNN,with a maximum improvement up to 5.2%,which proves the effectiveness of the proposed method.

Key words: Session-based recommendation, Recommendation system, Graph convolutional network, Attention mechanism, Over-smoothing

中图分类号: 

  • TP391
[1]SARWAR B,KARYPIS G,KONSTAN J,et al.Item-based collaborative filtering recommendation algorithms[C]//Procee-dings of the 10th International Conference on World Wide Web.2001:285-295.
[2]HIDASI B,KARATZOGLOU A,BALTRUNAS L,et al.Ses-sion-based recommendations with recurrent neural networks[J].arXiv:1511.06939,2015.
[3]KANG W C,MCAULEY J.Self-attentive sequential recommen-dation[C]//2018 IEEE International Conference on Data Mi-ning(ICDM).IEEE,2018:197-206.
[4]LI J,REN P,CHEN Z,et al.Neural attentive session-basedrecommendation[C]//Proceedings of the 2017 ACM on Confe-rence on Information and Knowledge Management.2017:1419-1428.
[5]WANG H,LIU G,LIU A,et al.DMRAN:A Hierarchical Fine-Grained Attention-Based Network for Recommendation[C]//IJCAI.2019:3698-3704.
[6]WANG M,REN P,MEI L,et al.A collaborative session-based recommendation approach with parallel memory modules[C]//Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval.2019:345-354.
[7]WANG S,HU L,WANG Y,et al.Modeling multi-purpose sessions for next-item recommendations via mixture-channel purpose routing networks[C]//International Joint Conference on Artificial Intelligence.International Joint Conferences on Artificial Intelligence.2019.
[8]LIU Q,ZENG Y,MOKHOSI R,et al.STAMP:short-term attention/memory priority model for session-based recommendation[C]//Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining.2018:1831-1839.
[9]LI Y,TARLOW D,BROCKSCHMIDT M,et al.Gated graph sequence neural networks[J].arXiv:1511.05493,2015.
[10]WU S,TANG Y,ZHU Y,et al.Session-based recommendation with graph neural networks[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2019:346-353.
[11]HU C,WANG Q,et al.Complex Structural Information for Session-based Recommendation with Graph Neural Networks[J].Computer Engineering,2022,48(5):82-90.
[12]WANG Z,WEI W,CONG G,et al.Global context enhancedgraph neural networks for session-based recommendation[C]//Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval.2020:169-178.
[13]DING H.Research on Optimization Algorithm of Graph Convolutional Neural Network[D].Beijing:University of Science and Technology,2022
[14]RENDLE S,FREUDENTHALER C,SCHMIDT-THIEME L.Factorizing personalized markov chains for next-basket recommendation[C]//Proceedings of the 19th International Confe-rence on World Wide Web.2010:811-820.
[15]JANNACH D,LUDEWIG M.When recurrent neural networks meet the neighborhood for session-based recommendation[C]//Proceedings of the Eleventh ACM Conference on Recommender Systems.2017:306-310.
[16]TAN Y K,XU X,LIU Y.Improved recurrent neural networks for session-based recommendations[C]//Proceedings of the 1st Workshop on Deep Learning for Recommender Systems.2016:17-22.
[17]VASWANI A,SHAZEER N,PARMAR N,et al.Attention isall you need[J].arXiv:1706,03762,2017.
[18]WANG S,HU L,WANG Y,et al.Sequential recommender systems:challenges,progress and prospects[J].arXiv:2001.04830,2019.
[19]WANG C,WANG C,WANG Z,et al.DeepDirect:Learning directions of social ties with edge-based network embedding[J].IEEE Transactions on Knowledge and Data Engineering,2018,31(12):2277-2291.
[20]WANG Z,YE X,WANG C,et al.Network embedding withcompletely-imbalanced labels[J].IEEE Transactions on Knowledge and Data Engineering,2020,33(11):3634-3647.
[21]WANG Z,YE X,WANG C,et al.RSDNE:Exploring relaxed similarity and dissimilarity from completely-imbalanced labels for network embedding[C]//Proceedings of the AAAI Confe-rence on Artificial Intelligence.2018.
[22]WANG C,WANG C,WANG Z,et al.Edge2vec:Edge-based social network embedding[J].ACM Transactions on Knowledge Discovery from Data(TKDD),2020,14(4):1-24.
[23]WANG Z,WANG J,GUO Y,et al.Zero-shot node classification with decomposed graph prototype network[C]//Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining.2021:1769-1779.
[24]QIU R,LI J,HUANG Z,et al.Rethinking the item order in session-based recommendation with graph neural networks[C]//Proceedings of the 28th ACM International Conference on Information and Knowledge Management.2019:579-588.
[25]KIPF T N,WELLING M.Semi-supervised classification withgraph convolutional networks[J].arXiv:1609.02907,2016.
[26]WU F,SOUZA A,ZHANG T,et al.Simplifying graph convolutional networks[C]//International Conference on Machine Learning.PMLR,2019:6861-6871.
[27]ZHU H,KONIUSZ P.Simple spectral graph convolution[C]//International Conference on Learning Representations.2020.
[28]WANG Z,WANG C,PEI J,et al.Multiple source detectionwithout knowing the underlying propagation model[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2017.
[29]SRIVASTAVA N,HINTON G,KRIZHEVSKY A,et al.Drop-out:a simple way to prevent neural networks from overfitting[J].The Journal of Machine Learning Research,2014,15(1):1929-1958.
[30]ZANGERLE E,PICHL M,GASSLER W,et al.# nowplaying music dataset:Extracting listening behavior from twitter[C]//Proceedings of the First International Workshop on Internet-scale Multimedia Management.2014:21-26.
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