Elamur Research⌕
☰
home / papers / 2020 / GCAN: Graph-aware Co-Attention Networks for Explainable Fake News Detection on Social Media

GCAN: Graph-aware Co-Attention Networks for Explainable Fake News Detection on Social Media

Paper · 2020 · Yi-Ju Lu and Cheng-Te Li

Summary

The paper tackles fake news detection on Twitter in a realistic but difficult setting. The source post is a short tweet, retweeters usually add no comments, and the social graph and diffusion tree are often hidden for privacy reasons. The authors propose GCAN (Graph-aware Co-Attention Networks), which uses three inputs: the tweet text, the ordered sequence of users who retweeted it, and features from those users' profiles.

The model has several parts:

  • A GRU encodes the source words.
  • CNN and GRU branches model how the retweet chain unfolds.
  • A GCN runs over a constructed user-interaction graph.
  • A dual co-attention mechanism links the source text with both the propagation and the interaction representations.

On real Twitter datasets, GCAN beats state-of-the-art baselines by about 16% accuracy on average. Its attention weights also point to suspicious retweeters and to the words they focus on. This matters because platforms or agencies can see why a story was flagged without needing long articles or user comments.

Key points

  • Detection runs under three constraints: short-text source tweets, no comment text, and no social or diffusion network structure. Only the retweet order and user profiles are used.
  • Experiments use the Twitter15 and Twitter16 datasets. The authors report an average accuracy gain of roughly 16% over strong prior models, and say GCAN also does well at early detection.
  • Ablation tests show that every component helps. The largest drops come from removing the dual co-attention, the propagation/interaction representations, or the source tweet itself.
  • In the case studies, suspicious users tend to be unverified, to have newer accounts and shorter profile descriptions, and to sit close to the original poster in the graph. Fake tweets draw attention to words like "breaking" and "strict", while real ones draw it to "confirmed" and "irrelevant".
  • The code is released at github.com/l852888/GCAN. The authors suggest GCAN could carry over to other short-text tasks such as sentiment analysis, hate speech detection, and popularity prediction.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓National Cheng Kung UniversityTainan · Taiwan3 works · 613 citations

📚 Shelves 6 Computer Science · 9,086 works Social Sciences · 744 works Information Systems · 523 works Sociology and Political Science · 221 works Physics and Astronomy · 25 works Statistical and Nonlinear Physics · 22 works

Cite

Cite as Yi-Ju Lu and Cheng-Te Li. 2020. GCAN: Graph-aware Co-Attention Networks for Explainable Fake News Detection on Social Media. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2020.acl-main.48

@inproceedings{lu2020gcan,
  title = {{GCAN: Graph-aware Co-Attention Networks for Explainable Fake News Detection on Social Media}},
  author = {Yi-Ju Lu and Cheng-Te Li},
  booktitle = {Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics},
  year = {2020},
  doi = {10.18653/v1/2020.acl-main.48},
  url = {https://doi.org/10.18653/v1/2020.acl-main.48}
}

References

33 references, 5 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Hunt Allcott and Matthew Gentzkow. 2017. Social media and fake news in the 2016 election. The Journal of Economic Perspectives, 31:211–235. find
  2. Carlos Castillo, Marcelo Mendoza, and Barbara Poblete. 2011. Information credibility on twitter. In Proceedings of the 20th International Conference on World Wide Web, WWW ’11, pages 675–684. find
  3. Meeyoung Cha, Wei Gao, and Cheng-Te Li. 2020. Detecting fake news in social media: An asia-pacific perspective. Commun. ACM, 63(4):68–71. find
  4. Junyoung Chung, Caglar Gulcehre, KyungHyun Cho, and Yoshua Bengio. 2014. Empirical evaluation of gated recurrent neural networks on sequence modeling. find
  5. Ming-Han Feng, Chin-Chi Hsu, Cheng-Te Li, Mi-Yen Yeh, and Shou-De Lin. 2019. Marine: Multi-relational network embeddings with relational proximity and node attributes. In The World Wide Web Conference, WWW ’19, pages 470–479. find
  6. Chuan Guo, Juan Cao, Xueyao Zhang, Kai Shu, and Miao Yu. 2019. Exploiting emotions for fake news detection on social media. CoRR, abs/1903.01728. — [arXiv:1903.01728](https://arxiv.org/abs/1903.01728) find
  7. Benjamin Horne and Sibel Adali. 2017. This just in: Fake news packs a lot in title, uses simpler, repetitive content in text body, more similar to satire than real news. In Proceedings of AAAI International Conference on Web and Social Media, pages 759–766. find
  8. Jyun-Yu Jiang, Cheng-Te Li, Yian Chen, and Wei Wang. 2018. Identifying users behind shared accounts in online streaming services. In The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval, SIGIR ’18, pages 65–74. find
  9. Thomas N. Kipf and Max Welling. 2017. Semi-Supervised Classification with Graph Convolutional Networks. In Proceedings of the 5th International Conference on Learning Representations, ICLR ’17. our card
  10. Haewoon Kwak, Changhyun Lee, Hosung Park, and Sue Moon. 2010. What is twitter, a social network or a news media? In Proceedings of the 19th International Conference on World Wide Web, WWW ’10, pages 591–600. find
  11. Sejeong Kwon, Meeyoung Cha, and Kyomin Jung. 2017. Rumor detection over varying time windows. PLOS ONE, 12(1):1–19. find
  12. Cheng-Te Li, Yu-Jen Lin, and Mi-Yen Yeh. 2018. Forecasting participants of information diffusion on social networks with its applications. Information Sciences, 422:432 – 446. find
  13. Yang Liu and Yi-Fang Wu. 2018. Early detection of fake news on social media through propagation path classification with recurrent and convolutional networks. In AAAI Conference on Artificial Intelligence, pages 254–261. find
  14. Jiasen Lu, Jianwei Yang, Dhruv Batra, and Devi Parikh. 2016. Hierarchical question-image co-attention for visual question answering. In Proceedings of the 30th International Conference on Neural Information Processing Systems, NIPS’16, pages 289–297. find
  15. Jing Ma, Wei Gao, Prasenjit Mitra, Sejeong Kwon, Bernard J. Jansen, Kam Fai Wong, and Meeyoung Cha. 2016. Detecting rumors from microblogs with recurrent neural networks. IJCAI International Joint Conference on Artificial Intelligence, pages 3818– 3824. find
  16. Jing Ma, Wei Gao, Zhongyu Wei, Yueming Lu, and Kam-Fai Wong. 2015. Detect rumors using time series of social context information on microblogging websites. In Proceedings of the 24th ACM International on Conference on Information and Knowledge Management, CIKM ’15, pages 1751–1754. find
  17. Jing Ma, Wei Gao, and Kam Fai Wong. 2017. Detect rumors in microblog posts using propagation structure via kernel learning. In ACL 2017 - 55th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference, pages 708– 717. our card
  18. Jing Ma, Wei Gao, and Kam-Fai Wong. 2018. Rumor detection on twitter with tree-structured recursive neural networks. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, pages 1980–1989. our card
  19. Kashyap Popat. 2017. Assessing the credibility of claims on the web. In Proceedings of the 26th International Conference on World Wide Web Companion, WWW ’17 Companion, pages 735–739. find
  20. Martin Potthast, Johannes Kiesel, Kevin Reinartz, Janek Bevendorff, and Benno Stein. 2018. A stylometric inquiry into hyperpartisan and fake news. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, ACL ’18, pages 231–240. our card
  21. Hannah Rashkin, Eunsol Choi, Jin Yea Jang, Svitlana Volkova, and Yejin Choi. 2017. Truth of varying shades: Analyzing language in fake news and political fact-checking. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 2931–2937. our card
  22. Julio C. S. Reis, André Correia, Fabrı́cio Murai, Adriano Veloso, and Fabrı́cio Benevenuto. 2019. Explainable machine learning for fake news detection. In Proceedings of the 10th ACM Conference on Web Science, WebSci ’19, pages 17–26. find
  23. Natali Ruchansky, Sungyong Seo, and Yan Liu. 2017. Csi: A hybrid deep model for fake news detection. In Proceedings of the 2017 ACM on Conference on Information and Knowledge Management, CIKM ’17, pages 797–806. find
  24. Justin Sampson, Fred Morstatter, Liang Wu, and Huan Liu. 2016. Leveraging the implicit structure within social media for emergent rumor detection. In Proceedings of the 25th ACM International on Conference on Information and Knowledge Management, CIKM ’16, pages 2377–2382. find
  25. Kai Shu, Limeng Cui, Suhang Wang, Dongwon Lee, and Huan Liu. 2019a. defend: Explainable fake news detection. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, KDD ’19, pages 395– 405. find
  26. Kai Shu, Amy Sliva, Suhang Wang, Jiliang Tang, and Huan Liu. 2017. Fake news detection on social media: A data mining perspective. SIGKDD Explor. Newsl., 19(1):22–36. find
  27. Kai Shu, Xinyi Zhou, Suhang Wang, Reza Zafarani, and Huan Liu. 2019b. The role of user profile for fake news detection. CoRR, abs/1904.13355. — [arXiv:1904.13355](https://arxiv.org/abs/1904.13355) find
  28. Pei-Chi Wang and Cheng-Te Li. 2019. Spotting terrorists by learning behavior-aware heterogeneous network embedding. In Proceedings of the 28th ACM International Conference on Information and Knowledge Management, CIKM ’19, pages 2097– 2100. find
  29. Yaqing Wang, Fenglong Ma, Zhiwei Jin, Ye Yuan, Guangxu Xun, Kishlay Jha, Lu Su, and Jing Gao. 2018. Eann: Event adversarial neural networks for multi-modal fake news detection. In Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery &\#38; Data Mining, KDD ’18, pages 849… find
  30. Rui Yan, Ian E.H. Yen, Cheng-Te Li, Shiqi Zhao, and Xiaohua Hu. 2015. Tackling the achilles heel of social networks: Influence propagation based language model smoothing. In Proceedings of the 24th International Conference on World Wide Web, WWW ’15, pages 1318–1328. find
  31. Fan Yang, Yang Liu, Xiaohui Yu, and Min Yang. 2012. Automatic detection of rumor on sina weibo. In Proceedings of the ACM SIGKDD Workshop on Mining Data Semantics, MDS ’12. find
  32. Yang Yang, Lei Zheng, Jiawei Zhang, Qingcai Cui, Zhoujun Li, and Philip S. Yu. 2018. Ti-cnn: Convolutional neural networks for fake news detection. find
  33. Zhe Zhao, Paul Resnick, and Qiaozhu Mei. 2015. Enquiring minds: Early detection of rumors in social media from enquiry posts. In Proceedings of the 24th International Conference on World Wide Web, WWW ’15, pages 1395–1405. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/lu-2020-gcan-graph-aware-co-attention/