计算机科学

计算机科学 ›› 2025, Vol. 52 ›› Issue (6A): 240900060-8.doi: 10.11896/jsjkx.240900060

• 大数据&数据科学 • 上一篇    下一篇

基于选择状态空间图神经网络的互联网应用用户画像分析

滕岷军1, 孙腾中1, 李彦辰1, 陈媛2, 宋沫飞3   

  1. 1 南京大数据集团有限公司 南京 211100
    2 南京市数据局 南京 210000
    3 东南大学计算机科学与工程学院 南京 211189
  • 出版日期:2025-06-16 发布日期:2025-06-12
  • 通讯作者: 宋沫飞(songmf@seu.edu.cn)
  • 作者简介:(tmjnjdsj@163.com)
  • 基金资助:
    国家自然科学基金(61906036)

Internet Application User Profiling Analysis Based on Selection State Space Graph Neural Network

TENG Minjun1, SUN Tengzhong1, LI Yanchen1, CHEN Yuan2, SONG Mofei3   

  1. 1 Nanjing Big Data Group Co.,Ltd.,Nanjing 211100,China
    2 Nanjing Data Bureau,Nanjing 210000,China
    3 School of Computer Science and Engineering,Southeast University,Nanjing 211189,China
  • Online:2025-06-16 Published:2025-06-12
  • About author:TENG Minjun,born in 1979.His main research interests include digital go-vernment and digital society.
    SONG Mofei,born in 1986,Ph.D,associate professor,Ph.D supervisor,is a member of CCF(No.H4907M).His main research interests include artificial intelligence and pattern recognition.
  • Supported by:
    National Natural Science Foundation of China(61906036).

PDF (PC)

36

摘要： 用户画像分析旨在深入挖掘用户在互联网应用中的偏好,对于推荐系统、个性化广告等众多实际应用具有十分重要的意义。近期的研究趋势是将用户及其互动对象视作图结构中的节点,从而将用户画像的构建转化为节点分类任务,并利用深度图神经网络技术来实现用户特征的提取。然而,这些研究往往未能充分考虑到不同用户间不同交互类型的差异性和时序关系,限制了用户画像分析的准确性。对此,提出了基于选择状态空间的图神经网络方法来进行用户画像分析,以同时捕捉图结构关系所蕴含的多用户对比和时序规律等上下文信息。为有效构建用户操作长序列数据的长程依赖关系,在图神经网络中引入状态空间模型,并结合基于注意力机制的节点优先级排列策略,以增强上下文感知推理,从而提高了用户性别和年龄等显式用户属性的预测性能。在两个真实的互联网APP数据集上进行了实验验证,结果证明了所提方法的有效性。

关键词: 用户画像, 图神经网络, 选择状态空间, 注意力机制, 时序分析

Abstract: User profile analysis aims to delve into users’ preferences in internet applications,which holds significant importance for various practical applications like recommendation systems and personalized advertising.Recent research trends consider users and their interactions as nodes in a graph structure,transforming user profile construction into a node classification task and utilizing deep graph neural network technology for user feature extraction.However,these studies often fail to fully consider the differences in interaction types among different users and their temporal relationships,thereby limiting the accuracy of user profile analysis.In light of this,this paper proposes a graph neural network method based on selected state space for user profile analysis to simultaneously capture context information such as multi-user comparisons and temporal patterns implied by graph structure relationships.To effectively model the long-range dependency relationships in user operation sequences,we introduce a state space model into the graph neural network and combine it with a node prioritization strategy based on attention mechanisms to enhance context-aware reasoning,thereby improving the predictive performance of explicit user attributes such as gender and age.Experimental validation on two real internet application datasets confirms the effectiveness of our proposed method.

Key words: User profile, Graph neural network, Selection state space, Attention mechanism, Timing analysis

中图分类号: 

  • TP311
[1]EKE C I,NORMAN A A,SHUIBL,et al.A survey of user profiling:State-of-the-art,challenges,and solutions[J].IEEE Access,2019,7:144907-144924.
[2]PURIFICATO E,BORATTO L,DE LUCA E W.User Mode-ling and User Profiling:A Comprehensive Survey[J].arXiv:2402.09660,2024.
[3]D′OCA S,HONG T.A data-mining approach to discover patterns of window opening and closing behavior in offices[J].Building and Environment,2014,82:726-739.
[4]VAN DAM J W,VAN DE VELDEN M.Online profiling andclustering of Facebook users[J].Decision Support Systems,2015,70:60-72.
[5]YAN S,ZHAO T,DENG J.Interaction-aware Hypergraph Neural Networks for User Profiling[C]//2022 IEEE 9th International Conference on Data Science and Advanced Analytics.IEEE,2022:1-10.
[6]LI M,HAN X,SHENG H,et al.A Novel RNN Model with Enhanced Behavior Semantic for Network User Profile[C]//2022 Tenth International Conference on Advanced Cloud and Big Data.IEEE,2022:190-193.
[7]XUAN H,LIU Y,LI B,et al.Knowledge enhancement for con-trastive multi-behavior recommendation[C]//Proceedings of the Sixteenth ACM International Conference on Web Search and Data Mining.2023:195-203.
[8]RAO D,YAROWSKY D,SHREEVATSA,et al.Classifying latent user attributes in twitter[C]//Proceedings of the 2nd International Workshop on Search and Mining User-generated Contents.2010:37-44.
[9]FERNANDEZ-LANVIN D,DE ANDRES-SUAREZ J,GONZA-LEZ-RODRIGUEZ M,et al.The dimension of age and gender as user model demographic factors for automatic personalization in e-commerce sites[J].Computer Standards & Interfaces,2018,59:1-9.
[10]CURA A,KÜÇÜK H,ERGEN E,et al.Driver profiling using long short-term memory(LSTM) and convolutional neural network(CNN) methods[J].IEEE Transactions on Intelligent Transportation Systems,2020,22(10):6572-6582.
[11]MEKRUKSAVANICH S,JITPATTANAKUL A.Convolutio-nal neural network and data augmentation for behavioral-based biometric user identification[C]//ICT Systems and Sustainability(CT4SD 2020).Springer Singapore,2021:753-761.
[12]YU Z,LIAN J,MAHMOODY A,et al.Adaptive User Modeling with Long and Short-Term Preferences for Personalized Recommendation[C]//IJCAI.2019:4213-4219.
[13]FAZIL M,SAH A K,ABULAISH M.Deepsbd:a deep neural network model with attention mechanism for socialbot detection[J].IEEE Transactions on Information Forensics and Security,2021,16:4211-4223.
[14]WANG X,SUN G,FANG X,et al.Modeling Spatio-temporal Neighbourhood for Personalized Point-of-interest Recommendation[C]//IJCAI.2022:3530-3536.
[15]ZHAO H,XIE J,WANG H.Co-learning Graph ConvolutionNetwork for Mobile User Profiling[J].Neural Processing Letters,2022,54(6):5299-5316.
[16]CHEN W,GU Y,REN Z,et al.Semi-supervised User Profiling with Heterogeneous Graph Attention Networks[C]//IJCAI.2019:2116-212.
[17]RAHIMI A,COHN T,BALDWIN T.Semi-supervised User Geolocation via Graph Convolutional Networks[C]//Proceedings of the 56th Annual Meeting of the Association for ComputationalLinguistics(Volume 1:Long Papers).2018:2009-2019.
[18]YAN Q,ZHANG Y,LIU Q,et al.Relation-aware heteroge-neous graph for user profiling[C]//Proceedings of the 30th ACM International Conference on Information & Knowledge Management.2021:3573-3577.
[19]QI Y,HU K,ZHANG B,et al.Trilateral spatiotemporal attention network for user behavior modeling in location-based search[C]//Proceedings of the 30th ACM International Conference on Information & Knowledge Management.2021:3373-3377.
[20]YUAN F,ZHANG G,KARATZOGLOU A,et al.One person,one model,one world:Learning continual user representation without forgetting[C]//Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval.2021:696-705.
[21]QI T,WU F,WU C,et al.News recommendation with candidate-aware user modeling[C]//Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval.2022:1917-1921.
[22]CHU Y W,HOSSEINALIPOUR S,TENORIO E,et al.Mitigating biases in student performance prediction via attention-based personalized federated learning[C]//Proceedings of the 31st ACM International Conference on Information & Knowledge Management.2022:3033-3042.
[23]LI M,HAN X,SHENG H,et al.A Novel RNN Model with Enhanced Behavior Semantic for Network User Profile[C]//2022 Tenth International Conference on Advanced Cloud and Big Data(CBD).IEEE,2022:190-193.
[24]AGARWAL P,SRIVASTAVA M,SINGHV,et al.Modelinguser behavior with interaction networks for spam detection[C]//Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval.2022:2437-2442.
[25]HAN J,LI W,CAI Z,et al.Multi-aggregator time-warping heterogeneous graph neural network for personalized micro-video recommendation[C]//Proceedings of the 31st ACM International Conference on Information & Knowledge Management.2022:676-685.
[26]YAN S,ZHAO T,DENG J.Interaction-aware Hypergraph Neural Networks for User Profiling[C]//2022 IEEE 9th International Conference on Data Science and Advanced Analytics.IEEE,2022:1-10.
[27]CHENG Z,HAN S,LIU F,et al.Multi-behavior recommendation with cascading graph convolution networks[C]//Procee-dings of the ACM Web Conference 2023.2023:1181-1189.
[28]ZHANG S,TONG H,XU J,et al.Graph convolutional net-works:a comprehensive review[J].Computational Social Networks,2019,6(1):1-23.
[29]KIPF T N,WELLING M.Semi-Supervised Classification withGraph Convolutional Networks[C]//International Conference on Learning Representations.2016.
[30]VELIČKOVIĆ P,CUCURULL G,CASANOVA A,et al.Graph Attention Networks[C]//International Conference on Learning Representations.2018.
[31]SCHLICHTKRULL M,KIPF T N,BLOEMP,et al.Modelingrelational data with graph convolutional networks[C]//The semantic web:15th international conference(ESWC 2018).Heraklion,Crete,Greece:Springer International Publishing,2018:593-607.
[32]HONG H,GUO H,LIN Y,et al.An attention-based graph neural network for heterogeneous structural learning[C]//Procee-dings of the AAAI Conference on Artificial Intelligence.2020:4132-4139.
[33]GU A,GOEL K,RÉ C.Efficiently modeling long sequences with structured state spaces[J].arXiv:2111.00396,2021.
[34]MEHTA H,GUPTA A,CUTKOSKY A,et al.Long range language modeling via gated state spaces[J].arXiv:2206.13947,2022.
[35]GU A,DAO T.Mamba:Linear-time sequence modeling with selective state spaces[J].arXiv:2312.00752,2023.
[36]LI K,LI X,WANG Y,et al.Videomamba:State space model for efficient video understanding[J].arXiv:2403.06977,2024.
[37]MA J,LI F,WANG B.U-mamba:Enhancing long-range de-pendency for biomedical image segmentation[J].arXiv:2401.04722,2024.
[38]LIU J,YU R,WANG Y,et al.Point mamba:A novel pointcloud backbone based on state space model with octree-based ordering strategy[J].arXiv:2403.06467,2024.
[39]XU K,HU W,LESKOVEC J,et al.How Powerful are Graph Neural Networks?[C]//International Conference on Learning Representations.2018.
[40]HU Z,DONG Y,WANG K,et al.Heterogeneous graph trans-former[C]//Proceedings of the Web Conference 2020.2020:2704-2710.
[41]WANG C,TSEPA O,MA J,et al.Graph-mamba:Towards long-range graph sequence modeling with selective state spaces[J].arXiv:2402.00789,2024.
[42]ABDELRAZEK M,PURIFICATO E,BORATTO L,et al.Fairup:A framework for fairness analysis of graph neural network-based user profiling models[C]//Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval.2023:3165-3169.
[43]PURIFICATO E,BORATTO L,DE LUCA E W.Toward a Re-sponsible Fairness Analysis:From Binary to Multiclass and Multigroup Assessment in Graph Neural Network-Based User Modeling Tasks[J].Minds and Machines,2024,34(3):33.
[44]JOULIN A,GRAVE E,BOJANOWSKI P,et al.Bag of tricks for efficient text classification[J].arXiv:1607.01759,2016.
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