融合背景知识的异构图虚假新闻检测方法研究
Research on Fake News Detection Method Using Heterogeneous Graph Fusion with Background Knowledge
DOI: 10.12677/csa.2024.143068, PDF,    科研立项经费支持
作者: 何 迈, 肖克晶*, 曹少中, 张 寒, 姜 丹：北京印刷学院信息工程学院，北京
关键词: 虚假新闻检测；异构图；图卷积网络模型；Fake News Detection； Heterogeneous Graph； GCN
摘要: 如今虚假新闻检测任务越来越受人们重视。本文考虑到不同的新闻具有涉及领域众多、隐含背景信息丰富的特点，提出利用新闻中的实体链接到领域广、信息全的维基百科，挖掘新闻潜在的背景信息与结构化三元组信息组成异构图，丰富新闻的表示。为了学习并更新建模后新闻异构图的特征向量，在图卷积网络的基础上，提出了一个基于语义距离的图卷积网络注意力模型DGAT (Distance Graph Attention Network, DGAT)。具体的，通过赋予异构图中不同类型节点不同的变化矩阵，将不同类型的节点映射到相同的公共空间中，解决了GCN模型不能直接应用在异构图上的局限。针对本文建模的新闻异构图特点，引入了基于新闻语义距离的注意力机制，以捕获融合了外部知识后，新闻与背景知识的语义一致性，最终输入分类器中进行虚假新闻检测。在公开数据集上进行的实验表明了本文方法的有效性。
Abstract: Nowadays, the task of fake news detection is receiving more and more attention. This article takes into account that different news has the characteristics of covering many fields and rich hidden background information. It is proposed to use the entities in the news to link to Wikipedia, which has a wide range of fields and complete information, to mine the potential background information of the news and form a heterogeneous graph with structured triplet information to enrich the representation of the news. In order to learn and update the modeled news heterogeneous graph feature vectors, an improved graph convolutional network model (GCN) and a Distance Graph Attention Network (DGAT) model are proposed. Specifically, by assigning different types of heterogeneous graphs to Different change matrices of nodes map different types of nodes into the same common space, solving the limitation that the GCN model cannot be directly applied to heterogeneous graphs. In view of the characteristics of the news heterogeneous graph modeled in this article, an attention mechanism based on news semantic distance is introduced to capture the semantic consistency of news and background knowledge after fusing external knowledge, and finally input it into the classifier to perform false recognition and news detection. Experiments on public datasets demonstrate the effectiveness of our method.
文章引用：何迈, 肖克晶, 曹少中, 张寒, 姜丹. 融合背景知识的异构图虚假新闻检测方法研究[J]. 计算机科学与应用, 2024, 14(3): 178-185. https://doi.org/10.12677/csa.2024.143068

参考文献

[1] Vosoughi, S., Roy, D. and Aral, S. (2018) The Spread of True and False News Online. Science, 359, 1146-1151. [Google Scholar] [CrossRef] [PubMed]
[2] Chen, Q., Zheng, Z. and Zhang (2020) Clinical Characteristics of 145 Patients with Corona Virus Disease 2019 (COVID-19) in Taizhou, Zhejiang, China. Infection, 48, 543-551. [Google Scholar] [CrossRef] [PubMed]
[3] Wang, Y., Qian, S., Hu, J., et al. (2020) Fake News Detection via Knowledge-Driven Multimodal Graph Convolutional Networks. Proceedings of the 2020 International Conference on Multimedia Retrieval, Dublin, 8-11 June 2020, 540-547. [Google Scholar] [CrossRef] 
[4] Hu, L.M., Yang, T.C., Zhang, L.H., Zhong, W.J., Tang, D.Y., Shi, C., Duan, N. and Zhou, M. (2021) Compare to the Knowledge: Graph Neural Fake News Detection with External Knowledge. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, Volume 1, 754-763.
[5] Bordes, A., Usunier, N., Garcia-Duran, A., Weston, J. and Yakhnenko, O. (2013) Translating Embeddings for Modeling Multi-Relational Data. NIPS Conference, Lake Tahoe, 5-8 December 2013, 2787-2795.
[6] Lu, Y.-J. and Li, C.-T. (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, 5-10 July 2020, 505-514.
[7] Jiang, S.Y., et al. (2019) User-Characteristic Enhanced Model for Fake News Detection in Social Media. 8th CCF International Conference, NLPCC 2019, Dunhuang, 9-14 October 2019, 634-646. [Google Scholar] [CrossRef] 
[8] Khoo, L.M.S., Chieu, H.L., Qian, Z. and Jiang, J. (2020) Interpretable Rumor Detection in Microblogs by Attending to User Interactions. Proceedings of the AAAI Conference on Artificial Intelligence, 34, 8783-8790. [Google Scholar] [CrossRef] 
[9] Bian, T., Xiao, X., Xu, T., Zhao, P., Huang, W., Rong, Y. and Huang, J. (2020) Rumor Detection on Social Media with Bi-Directional Graph Convolutional Networks. Proceedings of the AAAI Conference on Artificial Intelligence, 34, 549-556. [Google Scholar] [CrossRef] 
[10] Kang, Z.Z., et al. (2021) Fake News Detection with Heterogenous Deep Graph Convolutional Network. 25th Pacific-Asia Conference, PAKDD 2021, 11-14 May 2021, 408-420. [Google Scholar] [CrossRef] 
[11] Ma, J., Gao, W., Mitra, P., Kwon, S., Jansen, B.J., Wong, K.F. and Cha, M. (2016) Detecting Rumors from Microblogs with Recurrent Neural Networks.
[12] Shah, R., et al. (2019) SpotFake: A Multi-Modal Framework for Fake News Detection. IEEE 5th International Conference on Multimedia Big Data (BigMM), Singapore, 11-13 September 2019, 39-47.
[13] Zhou, X., Wu, J. and Zafarani, R. (2020) Similarity-Aware Multi-Modal Fake News Detection. 24th Pacific-Asia Conference, PAKDD 2020, Singapore, 11-14 May 2020, 354-367. [Google Scholar] [CrossRef] 
[14] Ferragina, P. and Scaiella, U. (2010) Tagme: On-the-Fly Annotation of Short Text Fragments (by Wikipedia Entities) Proceedings of the 19th ACM International Conference on Information and Knowledge Management, Toronto, 26-30 October 2010, 1625-1628. [Google Scholar] [CrossRef] 
[15] Kipf, T.N. and Welling, M. (2016) Semi-Supervised Classification with Graph Convolutional Networks.
[16] Rashkin, H., Choi, E., Jang, J.Y., Volkova, S. and Choi, Y. (2017) Truth of Varying Shades: Analyzing Language in Fake News and Political Fact-Checking. Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, Copenhagen, 7-11 September 2017, 2931-2937. [Google Scholar] [CrossRef] 
[17] Kim, Y. (2014) Convolutional Neural Networks for Sentence Classification. Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), Doha, 25-29 October 2014, 1746-1751. [Google Scholar] [CrossRef] 
[18] Hochreiter, S. and Schmidhuber, J. (1997) Long Short-Term Memory. Neural Computation, 9, 1735-1780. [Google Scholar] [CrossRef] [PubMed]
[19] Devlin, J., Chang, M.-W., Lee, K. and Toutanova, K. (2019) BERT: Pre-Training of Deep Bidirectional Transformers for Language Understanding. Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1, 4171-4186.
[20] Kipf, T.N. and Welling, M. (2017) Semi-Supervised Classification with Graph Convolutional Networks. International Conference on Learning Representations (ICLR), Toulon, 24-26 April 2017, 1121-1129.
[21] Aswani, A.V., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, Ł. and Polosukhin, I. (2017) Attention Is All You Need. Annual Conference on Neural Information Processing Systems 2017, Long Beach, 4-9 December 2017, 5998-6008.