计算机科学

计算机科学 ›› 2025, Vol. 52 ›› Issue (6A): 240600003-7.doi: 10.11896/jsjkx.240600003

• 大数据&数据科学 • 上一篇    下一篇

融合动态社会关系的下一个兴趣点推荐

蒋昊伦1, 朱金侠2, 孟祥福1   

  1. 1 辽宁工程技术大学电子与信息工程学院 辽宁 葫芦岛 125105
    2 辽宁工程技术大学软件学院 辽宁 葫芦岛 125105
  • 出版日期:2025-06-16 发布日期:2025-06-12
  • 通讯作者: 朱金侠(zjx15804291750@163.com)
  • 作者简介:(jhl20230805@163.com)
  • 基金资助:
    国家重点研发计划(2018YFB1402901);国家自然科学基金(61772249);辽宁省教育厅一般项目(LJ2019QL017)

Next Point of Interest Recommendation Incorporating Dynamic Social Relationships

JIANG Haolun1, ZHU Jinxia2, MENG Xiangfu1   

  1. 1 School of Electronic and Information Engineering,Liaoning Technical University,Huludao,Liaoning 125105,China
    2 School of Software,Liaoning Technical University,Huludao,Liaoning 125105,China
  • Online:2025-06-16 Published:2025-06-12
  • About author:JIANG Haolun,born in 1999.His main research interests include recommender systems and spatial data management.
    ZHU Jinxia,born in 1996.Her main research interests include recommender systems and so on.
  • Supported by:
    National Key Research and Development Program of China(2018YFB1402901),National Natural Science Foundation of China(61772249) and General Project of Liaoning Provincial Department of Education(LJ2019QL017).

PDF (PC)

81

摘要： 下一个兴趣点推荐现有的研究工作致力于通过整合用户偏好、序列行为以及时空上下文信息,优化推荐模型的准确性和实用性。尽管如此,当前的推荐策略仍面临着两大核心挑战:用户兴趣的动态性以及用户决策受其社会关系的影响。为了应对以上问题,提出一种融合动态社会关系的下一个兴趣点推荐模型。模型首先利用自注意力网络模拟用户偏好的动态变化,对序列信息、时空信息以及动态社会关系进行集成建模;其次设计了两个并行的长/短期通道,分别捕获用户的动态偏好以及上下文相关的动态社会关系;然后通过多头自注意力机制有效建模用户任意两个历史签到行为之间的长依赖关系,自适应地分配对下一个兴趣点的贡献值;最后在模型预测层利用注意力机制权衡用户长/短期偏好以及用户对兴趣点固有兴趣对用户决策的影响。在Gowalla和Brightkite这两个真实公开的数据集上进行实验,结果表明所提模型的推荐效果优于当前的下一个兴趣点推荐算法。

关键词: 下一个兴趣点推荐, 动态社会关系, 用户的动态偏好, 自注意力网络, 用户长/短期偏好

Abstract: The existing research on the next point-of-interest(POI) recommendation focuses on optimizing the accuracy and practicability of recommendation models by integrating user preferences,sequential behaviors,and spatio-temporal contextual information.However,current recommendation strategies still face two major challenges:1) the dynamic nature of user interests;2) the influence of users' social relations on their decision-making.To address these issues,a next POI recommendation model that integrates dynamic social relations is proposed.This model utilizes a self-attention network to simulate the dynamic changes in user preferences and performs integrated modeling of sequential information,spatio-temporal information,and dynamic social relations.Additionally,two parallel long-term and short-term channels are designed to capture users’ dynamic preferences and context-related dynamic social relations respectively.Through a multi-head self-attention mechanism,it effectively models the long-dependency relationship between any two historical check-in behaviors of users,adaptively allocating contribution values to the next POI.Finally,an attention mechanism is employed in the model's prediction layer to weigh the impact of users’ long-term and short-term preferences as well as their inherent interest in POI on their decision-making.Experiments on real-world and publicly available datasets from Gowalla and Brightkite demonstrate that the proposed model outperforms current next POI recommendation algorithms.

Key words: Next point-of-interest recommendations, Dynamic social relationships, User dynamic preferences, Self-attention networks, User long-term/short-term preferences

中图分类号: 

  • TP183
[1]CAI G Y,CHEN X Y,WANG S J.Novel Method Based on Graph Attention Aggregation for POI Recommendations[J].Computer Science,2022,49(S2):289-293.
[2]HUANG L W,MA Y T,LIU Y B,et al.Multi-Modal Bayesian Embedding for Point-of-Interest Recommendation on Location-Based Cyber-Physical-Social Networks [J].Future Generation Computer Systems,2020(108):111-1128.
[3]YANG C,SUN M S,XIN W Y,et al.A neural network approach to jointly modeling social networks and mobile trajectories[C]//Proceedings of the ACM Transactions International System.2017,36(4):1-28.
[4]ZHANG C M,ZHANG A R,ZHANG L,et al.A Graph Neural Networks-Based Learning Framework With Hyperbolic Embedding for Personalized Tag Recommendation[J].IEEE Access,2024(12):339-350.
[5]ZENG B Q,CHI J L,HONG P L,et al.Context-aware graph embedding with gate and attention for session-based recommendation[J].Neurocomputing,2024(574):127221.
[6]ZHAO P P,ZHU H F,LIU Y C,et al.Where to Go Next:ASpatio-Temporal Gated Network for Next POI Recommendation[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2019:5877-5884.
[7]XIAN X F,CHEN X J,ZHAO P P,et al.Context-aware personalized metric embedding for next POI recommendation[J].Computer Engineering & Science,2018,40(4):616-625.
[8]CHRISTOFORUDIS G,KEFALAS P,PAPADOPOULOS A,et al.Recommendation of Points-of-Interest using graph embeddings[C]//Proceedings of the Standard Authentication Algorithm.2018:31-40.
[9]LI Y,HANG Z Y,AN Z Y.Real-time preference-aware point-of-interest recommendation based on recurrent neural network[J].Computer Engineering an Design,2023,44(12):3772-3777.
[10]WU T H,SUN F Z,DONG J W,et al.Context-aware session recommendation based on recurrent neural networks[J].Comput Electr Eng.,2022,100:107916.
[11]CHAI R M,YIN C,MENG X F,et al.A recurrent neural network model based on spatial and temporal information for the next point of interest recommendation[J].CAAI Transactions on Intelligent Systems,2021,16(3):407-415.
[12]FENG J,LI Y,ZHANG C,et al.DeepMove:Predicting Human Mobility with Attentional Recurrent Networks[C]//Procee-dings of International Conference on World Wide Web.2018:1459-1468.
[13]YAN L,ZHAO P P,SHENG V S,et al.Social personalizedranking embedding for next poi recommendation[C]//International Conference on Web Information Systems Engineering.2017:91-105.
[14]GUO L,WEN Y F,LIU F G.Location perspective-based neighborhood-aware POI recommendation in location based social networks[J].Soft Computing,2019,23(22):11935-11945.
[15]HUANG L W,MA Y T,LIU Y B,et al.DAN-SNR:A Deep Attentive Network for Social-aware Next Point-of-interest Recommendation[C]//Proceedings of the ACM Transactions International System.2021:1-27.
[16]ADITYA G,JURE L.node2vec:Scalable feature learning fornetworks[C]//Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.2016,855-864.
[17]ZHOU C,BAI J Z,SONG J S,et al.Atrank:An attention-based user behavior modeling framework for recommendation[C]//Proceedings of AAAI Conference on Artificial Intelligence.2018,4564-4571.
[18]ASHISH V,NOAM S,NIKI P,et al.Attention is all you need.[C]//Proceedings of the Advances in Neural Information Processing Systems.2017:5998-6008.
[19]STEFFEN R,CHRISTOPH F,ZENO G,et al.BPR:Bayesian Personalized Ranking from Implicit Feedback.[C]//Proceedings of International Conference on Uncertainty in Artificial Intelligence.2009:452-461.
[20]LI H,YUE P,LI S C,et al.Spatio-temporal intention learning for recommendation of next point-of-interest[J].Geo Spatial Inf.Sci.,2024,27(2):384-397.
[21]BALAZS H,ALEXANDROS K,LINAS B,et al.Session-based recommendations with recurrent neural networks.[C]//Proceedings of the International Conference on Learning Representations.2015,1511:06939.
[22]HUANG L W,M Y T,WANG S B,et al.An Attention-based Spatiotemporal LSTM Network for Next POI Recommendation[C]//Proceedings of the IEEE Transactions International System.2019.
[23]OU J J,JIN H M,WANG X C,et al.STA-TCN:Spatial-temporal Attention over Temporal Convolutional Network for Next Point-of-interest Recommendation[J].ACM Transactions on Knowledge Discovery from Data,2023,17(9):1-19.
[24]KINGMA D,BA J.Adam:a method for stochastic optimization.[C]//Proceeding of the International Conference on Learning Representations.2015:1412-1427.
[25]ZHU J X,MENG X F,XING C Z,et al.Collaborative filtering recommendation approach fused with graph convolutional attention mechanism[J].CAAI Transactions on Intelligent Systems,2023,18(6):1295-1304.
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