Rumor Detection Model Based on Dynamic Propagation and Community Structure

doi:10.3778/j.issn.1002-8331.2306-0347

Abstract

Abstract: To address the insufficient utilization of time information, and to verify that the community structure features of rumor propagation can improve the performance of rumor detection models. Dy_PCRD (rumor detection model based on dynamic propagation and community structure) model is proposed which integrates dynamic propagation and community structure. On the one hand, GCN can extract structural features of rumor propagation, and on the other hand, topic communities are divided based on rumor content and propagation structure, and then a new attention calculation method is used to extract community structural features. They are inputted into temporal attention units to model their dynamic changes and classify the rumors. The experimental results on three public datasets show that under the same condition, its accuracy and other evaluation indicators have been improved compared to the baseline model, verifying the effectiveness of community structure features, dynamic features, and related attention calculation methods in improving the performance of rumor detection models.

Key words: rumor detection, dynamic graph, community structure, propagation, attention mechanism

摘要： 为解决现有谣言检测模型对时间信息利用不充分的问题，同时验证利用谣言传播的社区结构特征可以提高谣言检测模型的性能，提出一种融合动态传播和社区结构的社交媒体谣言检测模型Dy_PCRD（rumor detection model based on dynamic propagation and community structure），一方面使用图卷积网络提取谣言传播的结构特征，另一方面先根据谣言内容和传播结构划分话题社区，再使用一种新型的注意力计算方法提取谣言的社区结构特征，将二者分别输入时间注意力单元对其动态变化规律进行建模，最后基于所获得的嵌入表示对谣言进行分类。三个公开数据集上的实验结果表明，在相同条件下，相较于基线模型，其准确率及其他各评价指标均有所提升，验证了社区结构特征、动态性特征以及相关注意力计算方法对提升谣言检测模型性能的有效性。

关键词: 谣言检测, 动态图, 社区结构, 传播, 注意力机制

QIANG Zishan, GU Yijun. Rumor Detection Model Based on Dynamic Propagation and Community Structure[J]. Computer Engineering and Applications, 2024, 60(18): 198-207.

强子珊, 顾益军. 融合动态传播和社区结构的社交媒体谣言检测模型[J]. 计算机工程与应用, 2024, 60(18): 198-207.

References

[1] YANG J, PAN Y. COVID-19 rumor detection on social networks based on content information and user response[J]. Frontiers in Physics, 2021: 570.
[2] 刘政, 卫志华, 张韧弦. 基于卷积神经网络的谣言检测[J]. 计算机应用, 2017, 37(11): 3053-3056.
LIU Z, WEI Z H, ZHANG R X. Rumor detection based on convolutional neural network[J]. Journal of Computer Applications, 2017, 37(11): 3053-3056.
[3] 周丽娜, 谭励, 曹娟, 等. 基于卷积神经网络的食品安全领域谣言检测方法[J]. 计算机应用与软件, 2022, 39(3): 45-50.
ZHOU L N, TAN L, CAO J, et al. Rumor detection method in food safety field based on convolutional neural network[J]. Computer Applications and Software, 2022, 39(3): 45-50.
[4] GHANEM B, PONZETTO S P, ROSSO P, et al. Fakeflow: fake news detection by modeling the flow of affective information[C]//Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics, 2021: 679-689.
[5] ALKHODAIR S A, DING S H H, FUNG B C M, et al. Detecting breaking news rumors of emerging topics in social media[J]. Information Processing & Management, 2020, 57(2): 102018.
[6] MONTI F, FRASCA F, EYNARD D, et al. Fake news detection on social media using geometric deep learning[J]. arXiv:1902.06673, 2019.
[7] BIAN T, XIAO X, XU T, et al. Rumor detection on social media with bi-directional graph convolutional networks[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2020: 549-556.
[8] CUI W, SHANG M. KAGN: knowledge-powered attention and graph convolutional networks for social media rumor detection[J] Journal of Big Data, 2023: 45.
[9] WANG S, KONG Q, WANG Y, et al. Enhancing rumor detection in social media using dynamic propagation structures[C]//Proceedings of the 2019 IEEE International Conference on Intelligence and Security Informatics (ISI), 2019: 41-46.
[10] HUANG Q , ZHOU C, WU J, et al. Deep spatial-temporal structure learning for rumor detection on twitter[J]. Neural Computing and Applications, 2023, 35(18): 12995-13005.
[11] MA J, GAO W, WONG K F. Detect rumors in microblog posts using propagation structure via kernel learning[C]//Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, 2017: 708-717.
[12] KWON S, CHA M, JUNG K, et al. Prominent features of rumor propagation in online social media[C]//Proceedings of 2013 IEEE 13th International Conference on Data Mining, 2013: 1103-1108.
[13] BHATTACHARJEE U, SRIJITH P K, DESARKAR M S. Term specific tf-idf boosting for detection of rumours in social networks[C]//Proceedings of the 2019 11th International Conference on Communication Systems & Networks (COMSNETS), 2019: 726-731.
[14] MA J, GAO W, MITRA P, et al. Detecting rumors from microblogs with recurrent neural networks[C]//Proceedings of the International Joint Conference on Artificial Intelligence, 2016.
[15] WU L, LIU H. Tracing fake-news footprints: characterizing social media messages by how they propagate[C]//Proceedings of the Eleventh ACM International Conference on Web Search and Data Mining, 2018: 637-645.
[16] CHEN T, LI X, YIN H, et al. Call attention to rumors: deep attention based recurrent neural networks for early rumor detection[C]//Trends and Applications in Knowledge Discovery and Data Mining: PAKDD 2018 Workshops, BDASC, BDM, ML4Cyber, PAISI, DaMEMO, Melbourne, VIC, Australia, June 3, 2018. [S.l.]: Springer International Publishing, 2018: 40-52.
[17] WU K, YANG S, ZHU K Q. False rumors detection on sina weibo by propagation structures[C]//Proceedings of 2015 IEEE 31st International Conference on Data Engineering, 2015: 651-662.
[18] WANG Z, ZHANG W, TAN C W. On inferring rumor source for SIS model under multiple observations[C]//Proceedings of the 2015 IEEE International Conference on Digital Signal Processing (DSP), 2015: 755-759.
[19] ZHOU Y, WU C, ZHU Q, et al. Rumor source detection in networks based on the SEIR model[J]. IEEE Access, 2019, 7: 45240-45258.
[20] 薛海涛, 王莉, 杨延杰, 等. 基于用户传播网络与消息内容融合的谣言检测模型[J]. 计算机应用, 2021, 41(12): 3540-3545.
XUE H T, WANG L, YANG Y J, et al. Rumor detection model based on user propagation network and message content[J]. Journal of Computer Applications, 2021, 41(12): 3540-3545.
[21] MA J, GAO W, WONG K F. Rumor detection on twitter with tree-structured recursive neural networks[C]//Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 2018.
[22] SONG C, SHU K, WU B. Temporally evolving graph neural network for fake news detection[J]. Information Processing and Management, 2021, 58(6): 102712.
[23] WEI S, WU B, XIANG A, et al. DGTR: dynamic graph transformer for rumor detection[J]. Frontiers in Research Metrics and Analytics, 2023: 1055348.
[24] CHOI J, KO T, CHOI Y, et al. Dynamic graph convolutional networks with attention mechanism for rumor detection on social media[J]. PloS One, 2021, 16(8): e0256039.
[25] WOO S, PARK J, LEE J Y, et al. Cbam: convolutional block attention module[C]//Proceedings of the European Conference on Computer Vision (ECCV), 2018: 3-19.
[26] ZUBIAGA A, LIAKATA M, PROCTER R. Exploiting context for rumour detection in social media[C]//Proceedings of 9th International Conference on Social Informatics: (SocInfo 2017), Oxford, UK, September 13-15, 2017. [S.l.]: Springer International Publishing, 2017: 109-123.
[27] FAGBOHUNGBE O, QIAN L. Impact of learning rate on noise resistant property of deep learning models[J]. arXiv:2205.07856, 2022.
[28] BROCK A, DONAHUE J, SIMONYAN K. Large scale gan training for high fidelity natural image synthesis[C//Proceedings of the International Conference on Learning Representations, 2018.
[29] CAI G, WANG Y, HE L. Learning smooth representation for unsupervised domain adaptation[J]. arXiv:1905.10748, 2019.
[30] ZHU Y, LYU F, HU C, et al. Encoder-decoder architecture for supervised dynamic graph learning: a survey[J]. arXiv:2203.10480, 2022.
[31] LI M, LU S, ZHANG L, et al. A community detection method for social network based on community embedding[J]. IEEE Transactions on Computational Social Systems, 2021, 8(2): 308-318.
[32] WANG X, CUI P, WANG J, et al. Community preserving network embedding[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2022.