融合BERT和VGG模型多模态虚假新闻检测方法研究

曾江峰; 王蕊; 黎欣雨; 马霄

doi:10.13266/j.issn.2095-5472.2023.040

PDF(1889 KB)

知识管理论坛 ›› 2023, Vol. 8 ›› Issue (6) : 502-513. DOI: 10.13266/j.issn.2095-5472.2023.040

研究论文

融合BERT和VGG模型多模态虚假新闻检测方法研究

曾江峰 ¹ ,
王蕊 ¹ ,
黎欣雨 ² ,
马霄 ²

作者信息 +

Research on Multimodal Fake News Detection Method Based on BERT and VGG Models

Author information +

文章历史 +

摘要

[目的/意义] 旨在通过融合BERT和VGG模型解决当前虚假新闻泛滥、虚假新闻自动检测准确度较低、智能化较低问题。[方法/过程] 使用预训练模型BERT和VGG将新闻中的图文分离并转化为特征向量集，并进行特征融合，运用SVM模型设计分类器实现多模态虚假新闻检测识别。[结果/结论] 实证结果表明，实验数据集F1值达到93%,相较于单独使用BERT和VGG模型提升7%与9%，该方法具有较高的准确率和召回率，能够有效地检测虚假新闻。

Abstract

[Purpose/Significance] The aim is to solve the current problems of the proliferation of fake news, low accuracy and low intelligence of automatic fake news detection by integrating BERT and VGG models. [Method/Process] BERT and VGG models were uesd to separate the graphics and texts in the news and convert them into feature vector sets, and the feature fusion was carried out. The SVM model was used to design a classifier to achieve multi-modal fake news detection and identification. [Result/Conclusion] The empirical result shows that the F1 value of the experimental dataset reaches 93%, which is 7 percentage points and 9 percentage points higher than that of the BERT and VGG models alone, indicating that the combination of the two models has good detection accuracy and recall rate, and can effectively detect fake news.

导出引用

曾江峰 , 王蕊 , 黎欣雨 , 等. 融合BERT和VGG模型多模态虚假新闻检测方法研究[J]. 知识管理论坛. 2023, 8(6): 502-513 https://doi.org/10.13266/j.issn.2095-5472.2023.040

Jiangfeng Zeng , Rui Wang , Xinyu Li , et al. Research on Multimodal Fake News Detection Method Based on BERT and VGG Models[J]. Knowledge Management Forum. 2023, 8(6): 502-513 https://doi.org/10.13266/j.issn.2095-5472.2023.040

中图分类号： TP391

参考文献

列表( 原文顺序 | 文献年度倒序 | 文中引用次数倒序 ) 可视化分析

[1]

新华社.中国共产党第二十次全国代表大会在京开幕习近平代表第十九届中央委员会向大会作报告[EB/OL].[2023-01-22].http://www.gov.cn/xinwen/2022-10/16/content_5718884.htm.（Xinhua News Agency. Twentieth National Congress of the Communist Party of China opens in Beijing Xi Jinping reports to the Congress on behalf of the 19th Central Committee [EB/OL].[2023-01-22]. http://www.gov.cn/xinwen/2022-10/16/content_5718884.htm.）

本文引用 [1]

[2]	ALLCOTTH, GENTZKOW M. Social media and fake news in the 2016 election[J]. Journal of economic perspectives, 2017, 31(2): 211-36. 本文引用 [1]

[3]	刘赏,沈逸凡.基于新闻标题—正文差异性的虚假新闻检测方法[J].数据分析与知识发现,2023,7(2):97-107.（LIU S, SHEN Y F. Fake news detection method based on news title-text variability[J]. Data analysis and knowledge discovery,2023,7(2):97-107.）本文引用 [1]

[4]	VOSOUGHIS, ROY D, ARAL S. The spread of true and false news online[J]. science, 2018, 359(6380): 1146-1151. 本文引用 [1]

[5]	GUO C, CAO J, ZHANG X, et al. Dean: learning dual emotion for fake news detection on social media[J]. arXiv e-prints, 2019: arXiv: 1903.01728. 本文引用 [1]

[6]	RUCHANSKY N, SEOS, LIU Y. CSI: a hybrid deep model for fake news detection[C]//Proceedings of the 2017 ACM on conference on information and knowledge management. New York: ACM,2017:797-806. 本文引用 [1]

[7]	刁海伦,王树义,王楠.基于多主体的微博网络虚假信息的集中甄别方法研究[J].情报科学,2016,34(2):37-44.（DIAO H L,WANG S Y,WANG N. Research on centralised screening method of microblogging network false information based on multi-subjects[J]. Information science,2016,34(2):37-44.）本文引用 [1]

[8]	ZHOU P, HAN X, MORARIU VI, et al. Learning rich features for image manipulation detection[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. Salt Lake City: IEEE,2018:1053-1061. 本文引用 [1]

[9]	BOIDIDOU C, READOU K, PAPADOPOULOS S, et al. Verifying multimedia use at mediaeval 2015[M]//MediaEval 2015. Wurzen: CEUR-WS, 2015: 1436. 本文引用 [1]

[10]	JIN Z, CAO J, ZHANG Y, ET AL. Novel visual and statistical image features for microblogs news verification[J]. IEEE transactions on multimedia, 2016, 19(3): 598-608. 本文引用 [1]

[11]	LIU Y, WU Y F. Early detection of fake news on social media through propagation path classification with recurrent and convolutional networks[C]//Proceedings of the AAAI conference on artificial intelligence. New Orleans: AAAI, 2018: 354-361. 本文引用 [1]

[12]	MIAN A, KHAN S. Coronavirus: the spread of misinformation[J]. BMC medicine, 2020, 18(1): 1-2. 本文引用 [1]

[13]	汪超.基于多模态融合的虚假新闻检测算法研究[D].安徽:安徽理工大学,2022. (WANG C. Research on false news detection algorithm based on multimodal fusion[D]. Anhui: Anhui University of Science & Technology, 2022.) 本文引用 [1]

[14]	PENG Q, JUAN C, TIANYUN Y, et al. Exploiting multi-domain visual information for fake news detection[C]//2019 IEEE international conference on data mining (ICDM). Beijing: IEEE, 2019: 518-527. 本文引用 [1]

[15]	陶霄,朱焱,李春平.基于注意力与多模态混合融合的谣言检测方法[J].计算机工程,2021,47(12):71-77.（TAO X, ZHU Y, LI C R. Rumour detection method based on attention and multimodal hybrid fusion[J]. Computer engineering,2021,47(12):71-77.）本文引用 [1]

[16]	KALIYAR R K, GOSWAMI A, NARANG P, et al. FNDNet–A deep convolutional neural network for fake news detection[J].Cognitive systems research,2020,61:32-44. 本文引用 [1]

[17]	DEEPAK S, CHITTURI B. Deep neural approach to Fake-News identification[J]. Procedia computer science,2020,167:2236-2243. 本文引用 [1]

[18]	GOLDANI M H, SAFABAKHSH R, MOMTAZI S. Convolutional neural network with margin loss for fake news detection[J]. Information processing & management,2021, 58(1):102418. 本文引用 [1]

[19]	SAHOO S R, Gupta B B. Multiple features based approach for automatic fake news detection on social networks using deep learning[J] Applied soft computing,2021,100:106983. 本文引用 [1]

[20]	HAKAK S, ALAZAB M, KHAN S, et al. An ensemble machine learning approach through effective feature extraction to classify fake news[J] Future generation computer systems,2021,117:47-58. 本文引用 [1]

[21]	KHATTAR D, GOUD J S, GUPTA M, et al. Mvae: Multimodal variational autoencoder for fake news detection[C]//The world wide Web conference. New York: Association for Computing Machinery, 2019: 2915-2921. 本文引用 [1]

[22]	QI P, CAO J, YANG T, et al. Exploiting multi-domain visual information for fake news detection[C]//2019 IEEE international conference on data mining (ICDM). Beijing: IEEE, 2019: 518-527. 本文引用 [1]

[23]	亓鹏,曹娟,盛强.语义增强的多模态虚假新闻检测[J].计算机研究与发展,2021,58(7):1456-1465.（QI P, CAO J,SHENG Q. Semantic enhancement for multimodal fake news detection[J]. Journal of computer research and development,2021,58(7):1456-1465.）本文引用 [1]