Rumor Detection with Hierarchical Social Attention Network

Guo, Han; Cao, Juan; Zhang, Yazi; Guo, Junbo; Li, Jintao

doi:10.1145/3269206.3271709

Cited by 206 publications

(101 citation statements)

References 16 publications

Supporting

Mentioning

101

Contrasting

Order By: Relevance

“…e classifying accuracy of using account-based features alone is 72.6%. ese features have been used widely in subsequent methods, such as hierarchical neural networks [18]. Besides the classical features, Liu et al [19] considered the features of the credibility identification, diversity, and the relationship between profile location and event location.…”

Section: Related Workmentioning

confidence: 99%

Statistical Analysis of Dispelling Rumors on Sina Weibo

Deng

Wen

et al. 2020

Complexity

View full text Add to dashboard Cite

Analyzing the process and results of dispelling rumors is a prerequisite for designing an effective anti-rumor strategy. Current research on this subject focuses on the simulation experiments, short of empirical study. By using the False Information Publicity Results of Sina Weibo as the data source of empirical research, this article compares the typical features of rumor and anti-rumor accounts. Furthermore, taking COVID-19 as the target topic, distributions of the reported time, frequency, platform penalty levels, and diffusion parameters of rumors related to COVID-19 are given, and some interesting results are obtained.

show abstract

Section: Related Workmentioning

confidence: 99%

Statistical Analysis of Dispelling Rumors on Sina Weibo

Deng

Wen

et al. 2020

Complexity

View full text Add to dashboard Cite

show abstract

“…In addition to text features, multimodal features [18,19] and social features [48] are also exploited by recent studies. Besides, a few works study some new related tasks such as false event detection [30], false picture detection [33], and how to detect the false information [15] at an early stage. However, these methods mostly lack the property of explainability.…”

Section: False Information Detectionmentioning

confidence: 99%

“…For false information detection, the attention mechanism can help us distinguish the confusing input, select more key features, and improve the explainability of the model. Although existing false information detection methods have used the attention mechanism [15,36] , they only consider the importance of the input. In contrast, our proposed signed attention mechanism can jointly consider the importance and stance correlations between comments and posts, which would be beneficial to applications that require capturing richer semantic relationships.…”

Section: Attention Mechanismmentioning

confidence: 99%

QSAN: A Quantum-probability based Signed Attention Network for Explainable False Information Detection

Tian

Liu

Yang

et al. 2020

Proceedings of the 29th ACM International Conference on Information &Amp; Knowledge Management

View full text Add to dashboard Cite

False information detection on social media is challenging as it commonly requires tedious evidence-collecting but lacks available comparative information. Clues mined from user comments, as the wisdom of crowds, could be of considerable benefit to this task. However, it is non-trivial to capture the complex semantics from the contents and comments in consideration of their implicit correlations. Although deep neural networks have good expressive power, one major drawback is the lack of explainability. In this paper, we focus on how to learn from the post contents and related comments in social media to understand and detect the false information more effectively, with explainability. We thus propose a Quantum-probability based Signed Attention Network (QSAN) that integrates the quantum-driven text encoding and a novel signed attention mechanism in a unified framework. QSAN is not only able to distinguish important comments from the others, but also can exploit the conflicting social viewpoints in the comments to facilitate the detection. Moreover, QSAN is advantageous with its explainability in terms of transparency due to quantum physics meanings and the attention weights. Extensive experiments on real-world datasets show that our approach outperforms state-ofthe-art baselines and can provide different kinds of user comments to explain why a piece of information is detected as false.

show abstract

“…Ma et al in [29] discuss the same topic involving a novel approach to capture the temporal characteristics based on the time series of rumor's life cycle, for which time series modeling technique is applied to incorporate various social context information, while Han Guo et al [30] propose a novel hierarchical neural network combined with social information (HSA-BLSTM) for rumor detection and they test their model on two real-world datasets from Weibo and Twitter demonstrating outstanding performance in both rumor detection and early detection scenarios. Li et al [31] give another approach, the personalized influential topic search by proposing two random-walk based approaches in order to measure the influence of a topic on a query user. Moreover, Li et al in [32] studied the problem from another side, influence maximization; the aim is to find a limited number of users which can influence the maximum number of users in social networks.…”

Section: Related Workmentioning

confidence: 99%

Modeling a Rumor Propagation in Online Social Network: An Optimal Control Approach

Ghazzali

Bhih

Laaroussi

et al. 2020

Discrete Dynamics in Nature and Society

View full text Add to dashboard Cite

We propose to model the phenomenon of the spread of a rumor in social networks in this paper. From an existing SIR model, we manipulate a new one that is based on the model of cholera in order to take into account professional pages that specialize in spreading rumors. In the second part, we introduce a control strategy to fight against the diffusion of the rumor. Our main objective is to characterize the three optimal controls that minimize the number of spreader users, fake pages, and the corresponding costs. For that matter, using the maximum principle of Pontryagin, we prove the existence and we give characterization of our controls. Numerical simulations are given to concretize our approach.

show abstract

Rumor Detection with Hierarchical Social Attention Network

Cited by 206 publications

References 16 publications

Statistical Analysis of Dispelling Rumors on Sina Weibo

Statistical Analysis of Dispelling Rumors on Sina Weibo

QSAN: A Quantum-probability based Signed Attention Network for Explainable False Information Detection

Modeling a Rumor Propagation in Online Social Network: An Optimal Control Approach

Contact Info

Product

Resources

About