如今，社群媒體平台已經成為公眾接收和發布訊息、與他人交流的主要渠道，但由於平台的便利性，謠言也隨之四處傳播。而由於謠言的數量眾多，透過人工檢測很難立即發現並阻止謠言的傳播。因此，很多深度學習方法被用於自動的檢測謠言，如基於RNN和CNN的方法，但它們不能有效地獲得謠言傳播網絡的結構資訊，因此有人提出了一些基於RNN的樹結構模型，他們可以利用傳播網絡的結構，但它們的效率不好。基於GNN的謠言檢測方法可以有效地捕捉到謠言傳播網絡的結構資訊，如BiGCN和GACL是最先進的方法，但它們只使用含有文字內容的推文組成的傳播網絡進行謠言檢測，而放棄其他不含文字內容的推文。在本文中，我們提出了一個謠言檢測模型，它輸入了兩個圖，一個是由帶有文字的推文組成的文字圖，另一個是包含用戶資料、推文資訊和傳播特徵的完整傳播網絡的傳播圖。然後，我們使用RGCN編碼圖來學習自上而下和自下而上的傳播模式，並在每一層聚合後將兩種模式的資訊結合起來，以有效捕捉謠言的傳播結構。此外，我們還應用對比式學習來增強早期部分傳播圖的圖形表示，使在早期檢測上取得良好的效果。對於標籤稀疏的謠言檢測，我們應用了半監督學習方法，幫助所提出的方法取得良好的效果。最後，我們在兩個真實世界的數據集上進行了謠言檢測、早期謠言檢測、標籤稀疏的謠言檢測，以及魯棒性分析的實驗，證明我們提出的謠言檢測方法在大多數基准上優於最先進的方法。

Nowadays, social media platforms have become the main source for the public to receive and post message and communicate with others, but because of the convenience of the platforms, rumors are spread everywhere. And because of the large number of rumors, it is difficult to detect and stop the spread of rumors immediately through manual detection. Therefore, many deep learning methods are used to automatically detect rumors, such as RNN-based and CNN-based methods, but they cannot effectively obtain structural information of the rumor propagation network, so some tree structure models based on RNN have been proposed, which can use the structure of the propagation network, but they are not efficient enough. GNN-based rumor detection methods can effectively capture the structural information of rumor propagation networks, such as BiGCN and GACL are the state-of-the-art methods, but they only use the propagation network containing tweets with a text content for rumor detection, and drop the other tweets that did not contain text content. In this paper, we propose a rumor detection model that inputs two graphs, the text graph consisting of tweets with text, and the propagation graph with complete propagation network containing user profiles, tweet information, and propagation features. Then we use RGCN coded graphs to learn top-down and bottom-up propagation patterns and combine the information of both patterns after aggregation at each layer, to effectively capture rumor spreading structure. In addition, we apply contrastive learning to enhance the graph representation of early partial propagation graphs, which makes good results for early detection. For label sparse rumor detection, we apply a semi-supervised learning approach, which makes the proposed method have good results. Finally, we experiment on two real-world datasets for rumor detection, early rumor detection labeled sparse rumor detection, and robustness analysis, which demonstrate that our proposed method outperforms state-of-the-art methods on most benchmarks.

[1] Arazo, E., Ortego, D., Albert, P., O＇Connor, N. E., and McGuinness, K. Pseudolabeling and confirmation bias in deep semi-supervised learning. In 2020 International Joint Conference on Neural Networks (IJCNN) (2020), IEEE, pp. 1–8.
[2] Bian, T., Xiao, X., Xu, T., Zhao, P., Huang, W., Rong, Y., and Huang, J. Rumor detection on social media with bi-directional graph convolutional networks. In Proceedings of the AAAI conference on artificial intelligence (2020), vol. 34, pp. 549–556.
[3] Cascante-Bonilla, P., Tan, F., Qi, Y., and Ordonez, V. Curriculum labeling: Revisiting pseudo-labeling for semi-supervised learning. In Proceedings of the AAAI Conference on Artificial Intelligence (2021), vol. 35, pp. 6912–6920.
[4] Castillo, C., Mendoza, M., and Poblete, B. Information credibility on twitter. In Proceedings of the 20th international conference on World wide web (2011), pp. 675–684.
[5] Chen, T., Kornblith, S., Norouzi, M., and Hinton, G. A simple framework for contrastive learning of visual representations. In International conference on machine learning (2020), PMLR, pp. 1597–1607.
[6] Chen, T., Li, X., Yin, H., and Zhang, J. Call attention to rumors: Deep attention based recurrent neural networks for early rumor detection. In Pacific-Asia conference on knowledge discovery and data mining (2018), Springer, pp. 40–52.
[7] Chen, X., Fan, H., Girshick, R., and He, K. Improved baselines with momentum contrastive learning. arXiv preprint arXiv:2003.04297 (2020).
[8] Chen, X., and He, K. Exploring simple siamese representation learning. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (2021), pp. 15750–15758.
[9] Devlin, J., Chang, M.-W., Lee, K., and Toutanova, K. Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805 (2018).
[10] Ducci, F., Kraus, M., and Feuerriegel, S. Cascade-lstm: A tree-structured neural classifier for detecting misinformation cascades. In proceedings of the 26th ACM SIGKDD
international conference on Knowledge Discovery & Data Mining (2020), pp. 2666– 2676.
[11] Grill, J.-B., Strub, F., Altché, F., Tallec, C., Richemond, P., Buchatskaya, E., Doersch, C., Avila Pires, B., Guo, Z., Gheshlaghi Azar, M., et al. Bootstrap your own latent-a new approach to self-supervised learning. Advances in neural information processing systems 33 (2020), 21271–21284.
[12] Hadsell, R., Chopra, S., and LeCun, Y. Dimensionality reduction by learning an invariant mapping. In 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’06) (2006), vol. 2, IEEE, pp. 1735–1742.
[13] Han, S., Zhuang, F., He, Q., Shi, Z., and Ao, X. Energy model for rumor propagation on social networks. Physica A: Statistical Mechanics and its Applications 394 (2014), 99–109.
[14] Hassani, K., and Khasahmadi, A. H. Contrastive multi-view representation learning on graphs. In International Conference on Machine Learning (2020), PMLR, pp. 4116– 4126.
[15] Ioffe, S., and Szegedy, C. Batch normalization: Accelerating deep network training by reducing internal covariate shift. In International conference on machine learning (2015), PMLR, pp. 448–456.
[16] Kingma, D. P., and Ba, J. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014).
[17] Kwon, S., Cha, M., Jung, K., Chen, W., and Wang, Y. Prominent features of rumor propagation in online social media. In 2013 IEEE 13th international conference on data mining (2013), IEEE, pp. 1103–1108.
[18] Lee, D.-H., et al. Pseudo-label: The simple and efficient semi-supervised learning method for deep neural networks. In Workshop on challenges in representation learning, ICML (2013), vol. 3, p. 896.
[19] Liu, B., Sun, X., Meng, Q., Yang, X., Lee, Y., Cao, J., Luo, J., and Lee, R. K.-W. Nowhere to hide: Online rumor detection based on retweeting graph neural networks. IEEE Transactions on Neural Networks and Learning Systems (2022).
[20] Liu, X., Nourbakhsh, A., Li, Q., Fang, R., and Shah, S. Real-time rumor debunking on twitter. In Proceedings of the 24th ACM International on Conference on Information and Knowledge Management (2015), pp. 1867–1870.
[21] Liu, Y., and Wu, Y.-F. Early detection of fake news on social media through propagation path classification with recurrent and convolutional networks. In Proceedings of the AAAI conference on artificial intelligence (2018), vol. 32.
[22] Lu, Y.-J., and Li, C.-T. Gcan: Graph-aware co-attention networks for explainable fake news detection on social media. arXiv preprint arXiv:2004.11648 (2020).
[23] Ma, J., Gao, W., Mitra, P., Kwon, S., Jansen, B. J., Wong, K.-F., and Meeyoung, C. Detecting rumors from microblogs with recurrent neural networks. In The 25th International Joint Conference on Artificial Intelligence (2016), AAAI.
[24] Ma, J., Gao, W., Wei, Z., Lu, Y., and Wong, K.-F. 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 (2015), pp. 1751–1754.
[25] Ma, J., Gao, W., and Wong, K.-F. Detect rumors in microblog posts using propagation structure via kernel learning. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) (2017), vol. 1, pp. 708–717.
[26] Ma, J., Gao, W., and Wong, K.-F. Rumor detection on twitter with tree-structured recursive neural networks. Association for Computational Linguistics.
[27] Ma, J., Gao, W., and Wong, K.-F. Detect rumors on twitter by promoting information campaigns with generative adversarial learning. In The world wide Web conference (2019), pp. 3049–3055.
[28] Mikolov, T., Chen, K., Corrado, G., and Dean, J. Efficient estimation of word representations in vector space. arXiv preprint arXiv:1301.3781 (2013).
[29] Nguyen, D. Q., Vu, T., and Nguyen, A. T. Bertweet: A pre-trained language model for english tweets. arXiv preprint arXiv:2005.10200 (2020).
[30] Qiu, J., Chen, Q., Dong, Y., Zhang, J., Yang, H., Ding, M., Wang, K., and Tang, J. Gcc: Graph contrastive coding for graph neural network pre-training. In Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (2020), pp. 1150–1160.
[31] Rong, Y., Huang, W., Xu, T., and Huang, J. Dropedge: Towards deep graph convolutional networks on node classification. arXiv preprint arXiv:1907.10903 (2019).
[32] Schlichtkrull, M., Kipf, T. N., Bloem, P., Berg, R. v. d., Titov, I., and Welling, M. Modeling relational data with graph convolutional networks. In European semantic web conference (2018), Springer, pp. 593–607.
[33] Shi, W., Gong, Y., Ding, C., Tao, Z. M., and Zheng, N. Transductive semi-supervised deep learning using min-max features. In Proceedings of the European Conference on Computer Vision (ECCV) (2018), pp. 299–315.
[34] Silva, A., Han, Y., Luo, L., Karunasekera, S., and Leckie, C. Propagation2vec: Embedding partial propagation networks for explainable fake news early detection. Information Processing & Management 58, 5 (2021), 102618.
[35] Sun, T., Qian, Z., Dong, S., Li, P., and Zhu, Q. Rumor detection on social media with graph adversarial contrastive learning. In Proceedings of the ACM Web Conference 2022 (2022), pp. 2789–2797.
[36] Xie, Y., Xu, Z., Zhang, J., Wang, Z., and Ji, S. Self-supervised learning of graph neural networks: A unified review. IEEE Transactions on Pattern Analysis and Machine Intelligence (2022).
[37] Yang, F., Liu, Y., Yu, X., and Yang, M. Automatic detection of rumor on sina weibo. In Proceedings of the ACM SIGKDD workshop on mining data semantics (2012), pp. 1–7.
[38] Yang, X., Lyu, Y., Tian, T., Liu, Y., Liu, Y., and Zhang, X. Rumor detection on social media with graph structured adversarial learning. In Proceedings of the twenty-ninth international conference on international joint conferences on artificial intelligence (2021), pp. 1417–1423.
[39] Yang, X., Song, Z., King, I., and Xu, Z. A survey on deep semi-supervised learning. arXiv preprint arXiv:2103.00550 (2021).
[40] You, Y., Chen, T., Sui, Y., Chen, T., Wang, Z., and Shen, Y. Graph contrastive learning with augmentations. Advances in Neural Information Processing Systems 33 (2020), 5812–5823.
[41] Yu, F., Liu, Q., Wu, S., Wang, L., and Tan, T. Attention-based convolutional approach for misinformation identification from massive and noisy microblog posts. computers & security 83 (2019), 106–121.
[42] Yu, F., Liu, Q., Wu, S., Wang, L., Tan, T., et al. A convolutional approach for misinformation identification. In IJCAI (2017), pp. 3901–3907.
[43] Zhao, Z., Resnick, P., and Mei, Q. Enquiring minds: Early detection of rumors in social media from enquiry posts. In Proceedings of the 24th international conference on world wide web (2015), pp. 1395–1405.
[44] Zhu, Y., Xu, Y., Yu, F., Liu, Q., Wu, S., and Wang, L. Graph contrastive learning with adaptive augmentation. In Proceedings of the Web Conference 2021 (2021), pp. 2069– 2080.