[1]关昌珊,邴万龙,刘雅辉,等.基于图卷积网络的多特征融合谣言检测方法[J].郑州大学学报(工学版),2024,45(04):70-78.[doi:10.13705/ j.issn.1671-6833.2024.01.011]
　GUAN Changshan,BING Wanlong,LIU Yahui,et al.Multi-feature Fusion Rumor Detection Method Based on Graph Convolutional Network[J].Journal of Zhengzhou University (Engineering Science),2024,45(04):70-78.[doi:10.13705/ j.issn.1671-6833.2024.01.011]
点击复制

基于图卷积网络的多特征融合谣言检测方法()
分享到：

《郑州大学学报(工学版)》[ISSN:1671-6833/CN:41-1339/T]

卷:
45
期数:
2024年04期
页码:
70-78
栏目:
出版日期:
2024-06-16

文章信息/Info

Title:
Multi-feature Fusion Rumor Detection Method Based on Graph Convolutional Network
文章编号:
1671-6833(2024)04-0070-09
作者:
关昌珊;  邴万龙;  刘雅辉;  顾鹏飞;  马洪亮
石河子大学 信息科学与技术学院,新疆 石河子 832003
Author(s):
GUAN Changshan;  BING Wanlong;  LIU Yahui;  GU Pengfei;  MA Hongliang
School of Information Science and Technology, Shihezi University, Shihezi 832003, China
关键词:
谣言检测;  图卷积网络;  传播图;  传播用户;  特征融合
Keywords:
rumor detection;  graph convolutional network;  propagation graph;  propagation user;  feature fusion
分类号:
TP18
DOI:
10.13705/ j.issn.1671-6833.2024.01.011
文献标志码:
A
摘要:
目前,大部分谣言检测工作主要基于Twitter或新浪微博原文本内容、传播结构和传播文本内容进行谣言 检测,忽略了原文本特征与其他特征的有效融合,以及传播用户在谣言传播过程中的作用。针对以上问题,提出了 一种基于图卷积网络的多特征融合模型GCNs-BERT,模型同时融合了原文本特征、传播用户特征和传播结构特征。 首先,基于传播结构和传播用户构建传播图,将多个用户属性的组合作为传播节点特征;其次,利用多个图卷积网 络学习在不同用户属性组合的情况下传播图的表达,同时采用BERT模型学习原文本内容特征表达,最终与图卷 积网络学习的特征相融合用于检测谣言。利用公开的新浪微博数据集进行的大量实验表明:GCNs-BERT模型明显 优于基线方法。此外,在新冠疫情数据集上进行GCNs-BERT模型泛化能力实验,此数据集训练样本大小仅有新浪 微博数据集的1/5,仍然取得了92.5%的准确率,证明模型具有较好的泛化能力。
Abstract:
At present, most rumor detection work mainly based on the original text content, communication struc ture and communication text content of Twitter or Weibo. However, these methods ignored the effective integration of original text features with other features, as well as the role of propagating users in the process of rumor propaga tion. Aiming at the shortcomings of the existing work, a multi-feature fusion model GCNs-BERT based on graph convolutional network was proposed, which combined the features of the original text, the propagating user and the propagating structure. Firstly, a propagation graph was constructed based on the propagation structure and the prop agation users, and the combination of multiple user attributes was used as the propagation node feature. Then, mul tiple graph convolutional networks were used to learn the representation of the propagation graph with different user attribute combinations, and BERT model was used to learn the feature representation of the original text content. Finally, the fusion with the features learned by the graph convolutional network was used to detect rumors. A large number of experiments using publicly available Weibo data sets showed that the GCNs-BERT model was significant ly better than the baseline method. In addition, the generalization ability experiment of GCNs-BERT model was conducted on the novel coronavirus epidemic data set. The training sample size of this data set was only 1/5 of that of the public Weibo data set, and the accuracy rate was still 92.5%, which proved that the model had good gener alization ability.

参考文献/References:

[1] 中国互联网络信息中心. 第52次《中国互联网络发展状 况统计报告》[R/OL]. (2023-08-23)[2023-09-01]. ht tps:∥www.cnnic.net.cn/n4/2023/0828/c88-10829.html. 

China Internet Network Information Center. The 52nd China statistical report on internet development[R/OL]. (2023-08-23)[2023-09-01]. https:∥www. cnnic. net.cn/n4/2023/0828/c88-10829.html. 
[2] 梁兆君, 但志平, 罗衍潮, 等. 基于BERT模型的增 强混合神经网络的谣言检测[J]. 计算机应用与软件, 2021, 38(3): 147-152, 189. 
LIANG Z J, DAN Z P, LUO Y C, et al. Rumor detec tion of improved hybrid neural network based on BERT model[J]. Computer Applications and Software, 2021, 38(3)147-152, 189. 
[3] LI J W, NI S W, KAO H Y. Meet the truth: leverage objective facts and subjective views for interpretable rumor detection[EB/OL]. (2021-07-21)[2023-09-01]. https:∥arxiv.org/abs/2107.10747. 
[4] MA J, GAO W, MITRA P, et al. Detecting rumors from microblogs with recurrent neural networks[C]∥Twenty Fifth International Joint Conference on Artificial Intelli gence. New York: IJCAI, 2016: 3818-3824. 
[5] MA J, GAO W, WONG K F. Rumor detection on twitter with tree-structured recursive neural networks[C]∥Pro ceedings of the 56th Annual Meeting of the Association for Computational Linguistics.Stroudsburg:ACL,2018:1980-1989. 
[6] 胡斗, 卫玲蔚, 周薇, 等. 一种基于多关系传播树的 谣言检测方法[J]. 计算机研究与发展, 2021, 58 (7): 1395-1411. 
HU D, WEI L W, ZHOU W, et al. A rumor detection approach based on multi-relational propagation tree[J]. Journal of Computer Research and Development, 2021, 58(7): 1395-1411. 
[7] BIAN T, XIAO X, XU T Y, et al. Rumor detection on social media with bi-directional graph convolutional net works[J]. Proceedings of the AAAI Conference on Artifi cial Intelligence, 2020, 34(1): 549-556. 
[8] 杨延杰, 王莉, 王宇航. 融合源信息和门控图神经网 络的谣言检测研究[J]. 计算机研究与发展, 2021, 58 (7): 1412-1424. 
YANG Y J, WANG L, WANG Y H. Rumor detection based on source information and gating graph neural net work[J]. Journal of Computer Research and Develop ment, 2021, 58(7): 1412-1424. 
[9] BAI N, MENG F R, RUI X B, et al. Rumour detection based on graph convolutional neural net[J]. IEEE Ac cess, 2021, 9: 21686-21693. 
[10]孟青, 刘波, 张恒远, 等. 在线社交网络中群体影响 力的建模与分析[J]. 计算机学报, 2021, 44(6): 1064-1079. 
MENG Q, LIU B, ZHANG H Y, et al. Multi-relational group influence modeling and analysis in online social networks[J]. Chinese Journal of Computers, 2021, 44 (6): 1064-1079. 
[11]张铭泉, 周辉, 曹锦纲. 基于注意力机制的双BERT 有向情感文本分类研究[J]. 智能系统学报, 2022, 17 (6): 1220-1227. 
ZHANG M Q, ZHOU H, CAO J G. Dual BERT directed sentiment text classification based on attention mechanism [J]. CAAI Transactions on Intelligent Systems, 2022, 17(6): 1220-1227. 
[12] MA J, GAO W, WEI Z Y, et al. Detect rumors using time series of social context information on microblogging websites[C]∥Proceedings of the 24th ACM International on Conference on Information and Knowledge Manage ment. New York: ACM, 2015: 1751-1754. 
[13] LIU Y H, JIN X L, SHEN H W, et al. Do rumors dif fuse differently from non-rumors? a systematically empiri cal analysis in Sina Weibo for rumor identification[J]. Advances in Knowledge Discovery and Data Mining, 2017, 10234: 407-420. 
[14]WU K, YANG S, ZHU K Q. False rumors detection on Sina Weibo by propagation structures[C]∥2015 IEEE 31st International Conference on Data Engineering. Pisca taway: IEEE, 2015: 651-662. 
[15]WANG Y H, WANG L, YANG Y J, et al. SemSeq4FD: integrating global semantic relationship and local sequential order to enhance text representation for fake news de tection[J]. Expert Systems with Applications, 2021, 166: 114090. 
[16] MA J, GAO W, WONG K F. Detect rumors on Twitter by promoting information campaigns with generative ad versarial learning[C]∥WWW ′19: The World Wide Web Conference. New York: ACM, 2019: 3049-3055. 
[17] SHARMA K, QIAN F, JIANG H, et al. Combating fake news: a survey on identification and mitigation techniques [J]. ACM Transactions on Intelligent Systems and Tech nology, 10(3): 1-42. 
[18] PHAN H T, NGUYEN N T, HWANG D. Fake news de tection: a survey of graph neural network methods[J]. Applied Soft Computing, 2023, 139: 110235. 
[19] SONG C G, SHU K, WU B. Temporally evolving graph neural network for fake news detection[J]. Information Processing & Management, 2021, 58(6): 102712. 
[20] CASTILLO C, MENDOZA M, POBLETE B. Information credibility on Twitter[C]∥International World Wide Web Conference Committee (IW3C2). New York: ACM, 2011: 1-10. 
[21] LU Y J, LI C T. GCAN: graph-aware co-attention net works for explainable fake news detection on social media [EB/OL]. (2020-04-24)[2023-09-01]. https:∥ arxiv.org/abs/2004.11648. 
[22]刘雅辉, 靳小龙, 沈华伟, 等. 社交媒体中的谣言识别 研究综述[J]. 计算机学报, 2018, 41(7): 1536-1558. 
LIU Y H, JIN X L, SHEN H W, et al. A survey on ru mor identification over social media[J]. Chinese Journal of Computers, 2018, 41(7): 1536-1558. 
[23] YANG J, COUNTS S, MORRIS M R, et al. Microblog credibility perceptions: comparing the USA and China [C]∥Proceedings of the 2013 conference on Computer Supported Cooperative Work. New York: ACM, 2013: 575-586. 
[24] GONG S Z, SINNOTT R O, QI J Z, et al. Fake news detection through graph-based neural networks: a survey [EB/OL]. (2023-07-24)[2023-09-01]. https:∥ arxiv.org/abs/2307.12639. 
[25] LIU Y H, JIN X L, SHEN H W. Towards early identifi cation of online rumors based on long short-term memory networks[J]. Information Processing & Management, 2019, 56(4): 1457-1467. 
[26] DEVLIN J, CHANG M W, LEE K, et al. BERT: pre training of deep bidirectional transformers for language understanding[EB/OL]. (2019-05-24)[2023-09-01]. https:∥arxiv.org/abs/1810.04805.

更新日期/Last Update: 2024-06-14