计算机科学

计算机科学 ›› 2026, Vol. 53 ›› Issue (6): 304-314.doi: 10.11896/jsjkx.250400079

• 数据库&大数据&数据科学 • 上一篇    下一篇

基于因果干预缓解虚假关联的信息级联流行度预测

余柳, 李硕, 匡平, 周帆, 蒋涛   

  1. 电子科技大学信息与软件工程学院 成都 610054
  • 收稿日期:2025-04-17 修回日期:2025-07-09 出版日期:2026-06-15 发布日期:2026-06-09
  • 通讯作者: 匡平(kuangping@uestc.edu.cn)
  • 作者简介:(liu.yu@std.uestc.edu.cn)
  • 基金资助:
    四川省重点研发计划(2023YFG0114);四川省科技成果转化(2024ZHCG0031);成都市技术创新(2024YF0501231SN);四川省揭榜挂帅项目(2024YFCY0004);中央引导地方发展专项资金项目(2024ZYD0265)

Causal Intervention-based Mitigation of Spurious Correlations in Information Cascade PopularityPrediction

YU Liu, LI Shuo, KUANG Ping, ZHOU Fan, JIANG Tao   

  1. School of Information and Software Engineering,University of Electronic Science and Technology of China,Chengdu 610054,China
  • Received:2025-04-17 Revised:2025-07-09 Published:2026-06-15 Online:2026-06-09
  • About author:YU Liu,born in 1998,Ph.D candidate.Her main research interests include social network,large language model and responsible AI.
    KUANG Ping,born in 1977,Ph.D,professor,Ph.D supervisor.His main research interests include social network and computer vision.
  • Supported by:
    Sichuan Provincial Science and Technology Program Projects(2023YFG0114),Sichuan Provincial Science and Technology Program Projects(2024ZHCG0031),Science and Technology Program of Chengdu(2024YF0501231SN),Sichuan Province's Open Call for Proposals(2024YFCY0004)and Sichuan Province Central Leading Local Science and Technology Development Special Project(2024ZYD0265).

PDF (PC)

56

摘要： 信息传播中的级联流行度预测常受内部动态过程产生的虚假相关性影响。现有方法多假设级联数据满足独立同分布,但该假设在复杂传播环境中并不成立,易导致模型在分布外数据上的性能下降。为提升模型在异构数据下的稳定性与泛化能力,提出一种基于双重因果干预的级联预测方法——CCP。具体而言,级联内部干预通过随机修剪节点,打破观察级联规模与最终流行度之间的伪相关;级联间干预则引入结构相似级联的信息,以增加训练的多样性。CCP从因果视角解耦流行度与结构因素、观察时间等变量间的非因果关联,挖掘真正影响流行度的关键因子。在Weibo与APS两个数据集上,CCP相较现有先进方法CasCIFF,MSLE提升2%～5%,MAPE提升2%～3%,并在相同基线下展现出5%～7%的更优泛化性能。实验结果验证了CCP在复杂数据分布下的稳健性与有效性。

关键词: 信息级联, 因果干预, 流行度预测

Abstract: Information cascade popularity prediction is often affected by spurious correlations arising from internal cascade dynamics.Most existing methods assume that cascade data follows an independent and identically distributed pattern,which does not hold in real-world scenarios with complex diffusion processes.This mismatch leads to significant performance degradation on out-of-distribution data.To address this challenge and enhance model robustness and generalization under distributional shifts,this paper proposes CCP(Causal Cascade Prediction),a dual-intervention framework based on causal inference.Specifically,intra-cascade intervention randomly prunes nodes to break the misleading correlation between observed cascade size and final popularity,while inter-cascade intervention incorporates information from structurally similar cascades to introduce data diversity.CCP decouples popularity from non-causal factors such as structure and observation time,enabling the model to capture true causal drivers of information spread.Experimental results on the Weibo and APS datasets show that CCP outperforms the state-of-the-art CasCIFF method,achieving 2%~5% improvement in MSLE,and 2%~3% in MAPE,and demonstrates 5%~7% better generalization performance under the same baseline,validating its effectiveness in handling distributional shifts.

Key words: Information cascade, Causal intervention, Popularity prediction

中图分类号: 

  • TP391
[1]ZHOU F,XU X,TRAJCEVSKI G,et al.A survey of informationcascade analysis:Models,predictions,and recent advances[J].ACM Computing Surveys,2021,54(2):1-36.
[2]SHEN H,WANG D,SONG C,et al.Modeling and predictingpopularity dynamics via reinforced poisson processes[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2014.
[3]IRIBARREN J L,MORO E.Affinity paths and information diffusion in social networks[J].Social Networks,2011,33(2):134-142.
[4]MA S,FENG L,LAI C H.Mechanistic modelling of viral spreading on empirical social network and popularity prediction[J].Scientific Reports,2018,8(1):13126.
[5]CHENG Z,ZHONG T,ZHANG S.Survey of RecommenderSystems Based on Graph Learning [J].Journal of Computer Science,2022;49(9):1-13.
[6]HE S,TIAN H,LYU X.Edge popularity prediction based on social-driven propagation dynamics[J].IEEE Communications Letters,2017,21(5):1027-1030.
[7]BHARGAVA S,GHOSH D S.Analysis of feature reductiontechniques for online news popularity prediction[J].Smart Moves Journal IJOSCIENCE,2018,4:11-17.
[8]MISHRA S,RIZOIU M A,XIE L.Feature driven and pointprocess approaches for popularity prediction[C]//Proceedings of the 25th ACM International on Conference on Information and Knowledge Management.2016:1069-1078.
[9]CHENG J,ADAMIC L,DOW P A,et al.Can cascades be predicted?[C]//Proceedings of the 23rd International Conference on World Wide Web.2014:925-936.
[10]DALEY D J,VERE-JONES D.An introduction to the theory of point processes:volume I:elementary theory and methods[M].Springer Science & Business Media,2006.
[11]HAWKES A G.Spectra of some self-exciting and mutually exciting point processes[J].Biometrika,1971,58(1):83-90.
[12]YU L,XU X,TRAJCEVSKI G,et al.Transformer-enhancedHawkes process with decoupling training for information cascade prediction[J].Knowledge-Based Systems,2022,255:109740.
[13]YU L,XU X,ZHONG T,et al.Linking transformer to hawkes process for information cascade prediction(student abstract)[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2022:13103-13104.
[14]ZHAO Q,ERDOGDU M A,HE H Y,et al.Seismic:A self-exciting point process model for predicting tweet popularity[C]//Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.2015:1513-1522.
[15]DERAKHSHANI R,ESMAEELI H,AMIRI A.Phase II Monitoring of Poisson Regression Profiles in Multi-Stage Processes[J].International Journal of Reliability,Quality and Safety Engineering,2020,27(4):2050012.
[16]ZHAO Q,ZHANG Y,FENG X.Predicting information diffusion via deep temporalconvolutional networks[J].Information Systems,2022,108:102045.
[17]WANG J,ZHENG V W,LIU Z,et al.Topological recurrentneural network for diffusion prediction[C]//2017 IEEE International Conference on Data Mining(ICDM).IEEE,2017:475-484.
[18]GRAVES A.Long short-term memory[M]//Supervised Sequence Labelling with Recurrent Neural Networks.Berlin:Springer,2012:37-45.
[19]JAIN L,KATARYA R,SACHDEVA S.Opinion leaders for information diffusion using graph neural network in online social networks[J].ACM Transactions on the Web,2023,17(2):1-37.
[20]CAO Q,SHEN H,GAO J,et al.Popularity prediction on social platforms with coupled graph neural networks[C]//Proceedings of the 13th International Conference on Web Search and Data Mining.2020:70-78.
[21]CUI P,WANG J.Out-of-distribution(OOD) detection based on deep learning:A review[J].Electronics,2022,11(21):3500.
[22]ZHOU F,YU L,XU X,et al.Decoupling representation and regressor for long-tailed information cascade prediction[C]//Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval.2021:1875-1879.
[23]ZHOU F,XU X,ZHANG K,et al.Variational information diffusion for probabilistic cascades prediction[C]//IEEE INFOCOM 2020-IEEE Conference on Computer Communications.IEEE,2020:1618-1627.
[24]SUH B,HONG L,PIROLLI P,et al.Want to be retweeted? large scale analytics on factors impacting retweet in twitter network[C]//2010 IEEE Second International Conference on Social Computing.IEEE,2010:177-184.
[25]YU H,XIE L,SANNER S.The lifecyle of a youtube video:Phases,content and popularity[C]//Proceedings of the Internatio-nal AAAI Conference on Web and Social Media.2015:533-542.
[26]CARTA S,PODDA A S RECUPERO D R,et al.Popularity prediction of instagram posts[J].Information,2020,11(9):453.
[27]RIZOIU M A,XIE L,SANNER S,et al.Expecting to be hip:Hawkes intensity processes for social media popularity[C]//Proceedings of the 26th International Conference on World Wide Web.2017:735-744.
[28]GAO J,SHEN H,LIU S,et al.Modeling and predicting retweeting dynamics via a mixture process[C]//Proceedings of the 25th InternationalConference Companion on World Wide Web.2016:33-34.
[29]SHANG J,HUANG S,ZHANG D,et al.RNe2Vec:information diffusion popularity prediction based on repost network embedding[J].Computing,2021,103:271-289.
[30]TANG X,LIAO D,HUANG W,et al.Fully exploiting cascade graphs for real-time forwarding prediction[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2021:582-590.
[31]CHEN X,ZHOU F,ZHANG K,et al.Information diffusion prediction via recurrent cascades convolution[C]//2019 IEEE 35th International Conference on Data Engineering(ICDE).IEEE,2019:770-781.
[32]WANG Y,WANG X,JIA T.Ccasgnn:Collaborative cascadeprediction based on graph neural networks[C]//2022 IEEE 25th International Conference on Computer Supported Cooperative Work in Design(CSCWD).IEEE,2022:810-815.
[33]DENG Z,ZHENG X,TIAN H,et al.Deep causal learning:representation,discovery and inference[J].arXiv:2211.03374,2022.
[34]BAIARDI A,NAGHI A A.The value added of machine learning to causal inference:Evidence from revisited studies[J].The Econometrics Journal,2024,27(2):213-234.
[35]PEARL J.Models,reasoning and inference[M].Cambridge,UK:Cambridge University Press,2000:3.
[36]TRAN T Q,FUKUCHI K,AKIMOTO Y,et al.Unsupervisedcausal binary concepts discovery with vae for black-box model explanation[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2022:9614-9622.
[37]FEDER A,KEITH K A,MANZOOR E,et al.Causal inference in natural language processing:Estimation,prediction,interpretation and beyond[J].Transactions of the Association for Computational Linguistics,2022,10:1138-1158.
[38]WANG T,HUANG J,ZHANG H,et al.Visual commonsense r-cnn[C]//Proceedings of the IEEE/CVF Conference on Compu-ter Vision and Pattern Recognition.2020:10760-10770.
[39]ZHANG D,ZHANG H,TANG J,et al.Causal intervention for weakly-supervised semantic segmentation[J].Advances in Neural Information Processing Systems,2020,33:655-666.
[40]WANG W,LIN X,FENG F,et al.Causal representation lear-ning for out-of-distribution recommendation[C]//Proceedings of the ACM Web Conference 2022.2022:3562-3571.
[41]PEROZZI B,AL-RFOU R,SKIENA S.Deepwalk:Online lear-ning of social representations[C]//Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Disco-very and Data Mining.2014:701-710.
[42]GROVER A,LESKOVEC J.node2vec:Scalable feature learning for networks[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mi-ning.2016:855-864.
[43]VELIČKOVIĆ P,CUCURULL G,CASANOVA A,et al.Graph attention networks[J].arXiv:1710.10903,2017.
[44]SANDRYHAILA A,MOURA J M F.Discrete signal processing on graphs:Graph fourier transform[C]//2013 IEEE International Conference on Acoustics,Speech and Signal Processing.IEEE,2013:6167-6170.
[45]SHUMAN D I,NARANG S K,FROSSARD P,et al.The emerging field of signal processing on graphs:Extending high-dimensional data analysis to networks and other irregular domains[J].IEEE Signal Processing Magazine,2013,30(3):83-98.
[46]ZHU H,YUAN S,LIU X,et al.CasCIFF:A cross-domain information fusion framework tailored for cascade prediction in social networks[J].Knowledge-Based Systems,2024,303:112391.
[47]REICHENBACH H.The direction of time[M].University of California Press,1991.
[48]PAWLOWSKI N,COELHO DE CASTRO D,GLOCKER B.Deep structural causal models for tractable counterfactual inference[J].Advances in Neural Information Processing Systems,2020,33:857-869.
[49]HAGMAYER Y,SLOMAN S A,LAGNADO D A,et al.Causal reasoning through intervention[M]//Causal Learning:Psycho-logy,Philosophy,and Computation.2007:86-100.
[50]WEI Y,WANG X,NIE L,et al.Causal inference for knowledge graph based recommendation[J].IEEE Transactions on Know-ledge and Data Engineering,2022,35(11):11153-11164.
[51]NG I,ZHU S,FANG Z,et al.Masked gradient-based causalstructure learning[C]//Proceedings of the 2022 SIAM International Conference on Data Mining(SDM).Society for Industrial and Applied Mathematics,2022:424-432.
[52]CAO Q,SHEN H,CEN K,et al.Deephawkes:Bridging the gap between prediction and understanding of information cascades [C]//Proceedings of the 2017 ACM on Conference on Information and Knowledge Management.2017:1149-1158.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市两江新区洪湖西路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