Call Attention to Rumors: Deep Attention Based Recurrent Neural Networks for Early Rumor Detection

By Tong Chen, Lin Wu, Xue Li, Jun Zhang, Hongzhi Yin, Yang Wang. arXiv.org, 2017

323 citations recorded by the source corpus.

The proliferation of social media in communication and information dissemination has made it an ideal platform for spreading rumors. Automatically debunking rumors at their stage of diffusion is known as \textit{early rumor detection}, which refers to dealing with sequential posts regarding disputed factual claims with certain variations and highly textual duplication over time. Thus, identifying trending rumors demands an efficient yet flexible model that is able to capture long-range dependencies among postings and produce distinct representations for the accurate early detection. However, it is a challenging task to apply conventional classification algorithms to rumor detection in earliness since they rely on hand-crafted features which require intensive manual efforts in the case of large amount of posts. This paper presents a deep attention model on the basis of recurrent neural networks (RNN) to learn \textit{selectively} temporal hidden representations of sequential posts for identifying rumors. The proposed model delves soft-attention into the recurrence to simultaneously pool out distinct features with particular focus and produce hidden representations that capture contextual variations of relevant posts over time. Extensive experiments on real datasets collected from social media websites demonstrate that (1) the deep attention based RNN model outperforms state-of-the-arts that rely on hand-crafted features; (2) the introduction of soft attention mechanism can effectively distill relevant parts to rumors from original posts in advance; (3) the proposed method detects rumors more quickly and accurately than competitors.

Subfields: Artificial Intelligence, Sociology and Political Science

View Paper: https://www.semanticscholar.org/paper/83733442cbda97af26cd2e4e588b4ea07d1b2007

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Natali Ruchansky
2. Feng Qian
3. Karishma Sharma
4. He Jiang
5. Ming Zhang
6. Yan Liu
7. Wei Gao
8. Jing Ma

In Sociology and Political Science

1. Natali Ruchansky
2. Wei Gao
3. Kam-Fai Wong
4. Jing Ma
5. Feng Qian
6. Karishma Sharma
7. He Jiang
8. Ming Zhang

Citations in the corpus

• MVAE: Multimodal Variational Autoencoder for Fake News Detection (2019, methodology): “[5] incorporated attentionmechanism into recurrent neural networks (RNNs) to pool out distinct temporal-linguistic features with a particular focus.”
• Rumor Detection on Twitter with Tree-structured Recursive Neural Networks (2018, methodology): “…studies focused on text mining from sequential microblog streams using supervised models based on feature engineering (Castillo et al., 2011; Kwon et al., 2013; Liu et al., 2015; Ma et al., 2015), and more recently deep neural models (Ma et al., 2016; Chen et al., 2017; Ruchan-sky et al., 2017).”
• Fake News: A Survey of Research, Detection Methods, and Opportunities (2018, background): “…as the evolution of deceptive writing style, where Generative Adversarial Networks (GANs) [Wang et al. 2018] and Recurrent Neural Networks (RNNs) [Chen et al. 2017] have played or can play to their strengths; (3) verification efficiency, for example, by identifying check-worthy content and…”
• Cross-modal Ambiguity Learning for Multimodal Fake News Detection (2022, background): “CAR [7], which combines RNN with attention mechanism to capture relatively important textual information to detect text-only fake news.”
• Early Detection of Fake News on Social Media Through Propagation Path Classification with Recurrent and Convolutional Networks (2018, background): “(Chen et al. 2017) incorporate a soft-attention mechanism into the recurrent neural networks to pool out distinct temporal-linguistic features with a particular focus.”
• Hierarchical Multi-modal Contextual Attention Network for Fake News Detection (2021, background): “[4] propose a deep attention model on the basis of recurrent neural networks (RNN) to learn selectively temporal hidden representations of sequential posts for identifying fake news.”
• Combating Fake News (2019, background): “The deep attention used is nice for providing insights into interpretability but is not seen to improve detection accuracy in Chen et al. 2017, thereby is more useful for qualitative studies on fake news rather than detection.”