基于深度学习节点表示的谣言源定位方法

计算机工程与科学 ›› 2024, Vol. 46 ›› Issue (03): 545-559.

基于深度学习节点表示的谣言源定位方法

刘维,杨洁,罗佳莉，王赛威，陈崚

（扬州大学信息工程学院，江苏扬州 225000）

收稿日期:2023-09-01 修回日期:2023-10-28 接受日期:2024-03-25 出版日期:2024-03-25 发布日期:2024-03-18
基金资助:
国家自然科学基金(61971233,61702441)

Rumor source localization based on deep learning for node representation

LIU Wei,YANG Jie,LUO Jia-li,WANG Sai-wei,CHEN Ling

(School of Information Engineering,Yangzhou University,Yangzhou 225000,China)

Received:2023-09-01 Revised:2023-10-28 Accepted:2024-03-25 Online:2024-03-25 Published:2024-03-18

摘要/Abstract

摘要： 随着互联网的普及，网络上的信息以惊人的速度传播给公众。然而，由于级联效应，虚假信息和谣言同时也在迅速蔓延，对社会造成了巨大的危害。在社交网络上找到谣言的传播源头，对抑制谣言的传播起到至关重要的作用。传统的谣言源定位方法大多未能够融合多源特征且定位准确率还需进一步提高，因此，提出一种基于深度学习的谣言源定位方法，根据观测受谣言影响的节点多源特征来识别谣言源。首先，根据节点与观测节点之间的影响力相似度得到节点的影响力向量。然后，利用自编码网络对节点的影响力向量进行编码，得到包含节点信息、扩散路径和传播时间信息在内的节点的新的嵌入表示。最后，根据节点新的影响力向量计算节点为谣言源的概率，以定位谣言源。在2个模拟网络和4个真实网络上的实验结果表明，与其他方法相比，所提方法能够以更快的速度定位谣言源，且谣言源定位的准确率提升了25%以上。

关键词: 社交网络, 节点表示, 谣言源；多谣言源定位

Abstract: With the popularity of the Internet, information on the web is spreading to the public at an astonishing speed. However, false information and rumors are also rapidly spreading due to the cascade effect, causing great harm to society. Finding the source of rumor spread on social networks plays a crucial role in suppressing the spread of rumors. Most of the traditional rumor source localization methods fail to integrate multi-source features and the accuracy of localization still needs to be further improved. Therefore, this paper proposes a deep learning-based rumor source localization method that identifies the rumor source based on multi-source features observed in nodes affected by rumors. This method first obtains the influence vectors of nodes based on the similarity of influence between the node and observed nodes. Then, it uses autoencoder networks to encode the influence vectors of nodes, obtaining new embedding representations that contain node information, diffusion paths, and propagation time information. Finally, it calculates the probability of nodes being the source based on their new influence vectors to locate the rumor source. Experimental results on two simulated datasets and four real datasets show that compared with other methods, the proposed method can locate the rumor source at a faster speed and improve the accuracy of rumor source localization by more than 25%.

Key words: social network, node representation, rumor source, multi-rumor source localization

刘维, 杨洁, 罗佳莉, 王赛威, 陈崚. 基于深度学习节点表示的谣言源定位方法[J]. 计算机工程与科学, 2024, 46(03): 545-559.

LIU Wei, YANG Jie, LUO Jia-li, WANG Sai-wei, CHEN Ling. Rumor source localization based on deep learning for node representation[J]. Computer Engineering & Science, 2024, 46(03): 545-559.

参考文献［30］

［1］	韩雪明,贾彩燕,李轩涯,等.传播树结构结点及路径双注意力谣言检测模型［J］.计算机科学,2023,50(4):22-31.
	Han Xue-ming, Jia Cai-yan, Li Xuan-ya,et al. Dual- attention network model on propagation tree structures for rumor detection［J］. Computer Science, 2023, 50(4):22-31.
［2］	Shah D,Zaman T.Detecting sources of computer viruses in networks:Theory and experiment［J］. ACM SIGMETRICS Evaluation Review,2010,38(1):203-214.
［3］	Shah D,Zaman T.Rumor centrality:A universal source detector［J］. ACM SIGMETRICS Performance Evaluation Review,2012,40(1):199-210.
［4］	Luo W,Tay W P,Leng M,et al.On the universality of the Jordan center for estimating the rumor source in a social network［C］∥Proc of 2015 IEEE International Conference on Digital Signal Processing,2015:760-764.
［5］	Zang W,Zhang P,Zhou C,et al.Locating multiple sources in social networks under the SIR model:A divide-and-conquer approach［J］.Journal of Computational Science,2015,10:278-287.
［6］	Kalvit A,Borkar V S,Karamchandani N.Stochastic approximation algorithms for rumor source inference on graphs［J］.Performance Evaluation,2019,132:1-20.
［7］	Ding Y,Cui X,Wang H,et al.PRIA:A multi-source recognition method based on partial observation in SIR model［J］.Mobile Networks and Applications,2021,26:1514-1522.
［8］	Wang Z,Sun C,Rui X,et al.Localization of multiple diffusion sources based on overlapping community detection［J］.Knowledge-Based Systems,2021,226:106613.
［9］	Nguyen D T,Nguyen N P,Thai M T.Sources of misinformation in online social networks:Who to suspect?［C］∥Proc of IEEE Military Communications Conference,2012:1-6.
［10］	Zhang Z,Yue K,Sun Z,et al.Locating sources in online social networks via random walk［C］∥Proc of 2017 IEEE International Congress on Big Data,2017:337-343.
［11］	Dong M,Zheng B,Li G,et al.Wavefront-based multiple rumor sources identification by multi-task learning［J］.IEEE Transactions on Emerging Topics in Computational Intelligence,2022,6(5):1068-1078.
［12］	Dawkins Q E,Li T,Xu H.Diffusion source identification on networks with statistical confidence［C］∥Proc of International Conference on Machine Learning,2021:2500-2509.
［13］	Zhu K, Ying L.Information source detection in the SIR model:A sample-path-based approach［J］.IEEE/ACM Transactions on Networking,2014,24(1):408-421.
［14］	Jog V,Loh P L.Persistence of centrality in random growing trees［J］.Random Structures & Algorithms,2018,52(1):136-157.
［15］	Kouzy R,Abi Jaoude J,Kraitem A,et al.Coronavirus goes viral:Quantifying the COVID-19 misinformation epidemic on Twitter［J］.Cureus,2020,12(3):1-8.
［16］	Hu Z L, Shen Z,Cao S,et al.Locating multiple diffusion sources in time varying networks from sparse observations［J］.Scientific Reports,2018,8(1):2685.
［17］	Zhu K, Chen Z,Ying L.Locating the contagion source in networks with partial timestamps［J］.Data Mining and Knowledge Discovery,2016,30:1217-1248.
［18］	Khim J,Loh P L.Confidence sets for the source of a diffusion in regular trees［J］.IEEE Transactions on Network Science and Engineering,2016,4(1):27-40.
［19］	Li W M, Guo C, Liu Y X, et al. Rumor source localization in social networks based on infection potential energy［J］. Information Sciences:An International Journal, 2023, 634(C):172-188.
［20］	Qiu L Q, Sai S Q, Wei M J. BPSL: A new rumor source location algorithm based on the time-stamp back propagation in social networks［J］. Applied Intelligence, 2022,52(8):8603-8615.
［21］	Shu X,Yu B,Ruan Z,et al.Information source estimation with multi-channel graph neural network［M］∥Graph Data Mining:Algorithm,Security and Application.Singapore:Springer, 2021.
［22］	Chen X,Hu X,Wang C.Approximation for the minimum cost doubly resolving set problem［J］.Theoretical Computer Science,2016,609:526-543.
［23］	Leskovec J,Mcauley J.Learning to discover social circles in ego networks［J］.Advances in Neural Information Proces- sing Systems,2012,25(1):1-8.
［24］	Leskovec J,Huttenlocher D,Kleinberg J.Predicting positive and negative links in online social networks［C］∥Proc of the 19th International Conference on World Wide Web,2010:641-650.
［25］	Wang J W, Rong L L. Cascade-based attack vulnerability on the US power grid［J］. Safety Science, 2009, 47(10):1332-1336.
［26］	Asim Y, Niazi M A, Raza B, et al. Personal vs. know-how contacts: Which matter more in wiki elections?［J］. Complex Adaptive Systems Modeling, 2018, 6(1):1-19.
［27］	Ali S S,Anwar T,Rastogi A,et al.EPA:Exoneration and prominence based age for infection source identification［C］∥Proc of the 28th ACM International Conference on Information and Knowledge Management,2019:891-900.
［28］	Lokhov A Y,Mézard M,Ohta H,et al.Inferring the origin of an epidemic with a dynamic message-passing algorithm［J］.Physical Review E,2014,90(1):012801.
［29］	Antulov-Fantulin N,Lancic A,muc T,et al.Identification of patient zero in static and temporal networks:Robustness and limitations［J］.Physical Review Letters,2015,114(24):248701.
［30］	Jiang J,Wen S,Yu S,et al.Identifying propagation sources in networks:State-of-the-art and comparative studies［J］.IEEE Communications Surveys & Tutorials,2016,19(1):465-481.