Rumor Detection on Twitter with Tree-structured Recursive Neural Networks

By Jing Ma, Wei Gao, Kam-Fai Wong. Annual Meeting of the Association for Computational Linguistics, 2018

596 citations recorded by the source corpus.

Automatic rumor detection is technically very challenging. In this work, we try to learn discriminative features from tweets content by following their non-sequential propagation structure and generate more powerful representations for identifying different type of rumors. We propose two recursive neural models based on a bottom-up and a top-down tree-structured neural networks for rumor representation learning and classification, which naturally conform to the propagation layout of tweets. Results on two public Twitter datasets demonstrate that our recursive neural models 1) achieve much better performance than state-of-the-art approaches; 2) demonstrate superior capacity on detecting rumors at very early stage.

Subfields: Artificial Intelligence, Information Systems, Sociology and Political Science

View Paper: https://www.semanticscholar.org/paper/8594d9d30039065f542ef9db62d1a840cf4c3562

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. Wenbing Huang
2. Tingyang Xu
3. Xinyi Zhou
4. Junzhou Huang
5. Peilin Zhao
6. Yu Rong
7. Juan Cao
8. Xi Xiao

In Information Systems

1. Xinyi Zhou
2. Kai Shu
3. Juan Cao
4. Peilin Zhao
5. Vasudeva Varma
6. Feng Qian
7. Karishma Sharma
8. He Jiang

In Sociology and Political Science

1. Xinyi Zhou
2. Kai Shu
3. Yi-Ju Lu
4. Cheng-Te Li
5. Wenbing Huang
6. Juan Cao
7. Peilin Zhao
8. Xi Xiao

Citations in the corpus

• DTCA: Decision Tree-based Co-Attention Networks for Explainable Claim Verification (2020, methodology): “…of credibility-indicative features around semantics (Ma et al., 2018b; Khattar et al., 2019; Wu et al., 2020), emotions (Ajao et al., 2019), stances (Ma et al., 2018a; Kochkina et al., 2018; Wu et al., 2019), write styles (Potthast et al., 2018; Gr¨ondahl and Asokan, 2019), and source credibility…”
• A Survey of Fake News (2020, methodology): “The model (parameters) can be trained (estimated) by minimizing some cost function, such as squared error [Ma et al. 2018] and cross-entropy [Zhang et al. 2018], using some optimization algorithm, such stochastic gradient descent (SGD) [Wang et al. 2018], Adam [Kingma and Ba 2014], and the…”
• Rumor Detection on Social Media with Bi-Directional Graph Convolutional Networks (2020, methodology): “Many deep learning models such as Long Short Term Memory (LSTM), Gated Recurrent Unit (GRU), and Recursive Neural Networks (RvNN) (Ma et al. 2016; Ma, Gao, and Wong 2018) are employed since they are capable to learn sequential features from rumor propagation along time.”
• GCAN: Graph-aware Co-Attention Networks for Explainable Fake News Detection on Social Media (2020, background): “Third, some studies (Ma et al., 2018) consider that the pathways of information cascade (i.e., retweets) in the social network are useful for classifying misinformation, and thus learn the representations of the tree-based propagation structures.”
• Mining Dual Emotion for Fake News Detection (2019, background): “To tackle this issue, researchers have been devoted to developing automatic methods to detect fake news (i.e., designing a classifier to judge a given news piece as real or fake) by leveraging signals from text [5, 32, 34], images [20, 35], or social contexts [17, 24, 26– 28, 39, 40].”
• Combating Fake News (2019, methodology): “[65] not only significantly reduce the computation costs and time but also manage to improve over the detection accuracy obtained by the tree kernel methods [63].”
• MVAE: Multimodal Variational Autoencoder for Fake News Detection (2019, methodology): “Note that fair comparisons with methods proposed in [13, 15] are not possible since they use addition information like tweet propagation data.”