摘要
社交媒体改变了人们获取信息的方式,但也助长了错误信息在网络上的产生和传播。如何准确且快速地检测出社交媒体上的错误信息以净化网络环境,是一个重要的研究议题。本文从信息生态理论出发,分别从内容、用户和传播三个角度来阐述当前错误信息检测所关注的问题以及对应的检测方法,对近年来国内外的相关研究成果进行了系统梳理。现有的检测方法已经利用深度学习等技术取得了较好的检测结果。但是,由于错误信息爆发初期相关数据较少,有关早期检测的研究尚不多见;能够实现有效迁移和预训练的大规模基准数据集仍有待构建;从用户入手的信息挖掘有待进一步深入研究。
Social media has dramatically improved the efficiency of information access, but it has also contributed to the generation and dissemination of misinformation on the Internet. The accurate and quick detection of misinformation to improve the online information environment is an important issue. Inspired by the Information Ecology Theory, this paper expounds on the current problems and relevant detection methods of misinformation from the three perspectives of content,users, and dissemination. Existing detection methods have achieved state-of-the-art results by deep learning methods. However, because of the lack of relevant data in the early stages, studies on the early detection of misinformation are still rare.Additionally, large-scale benchmark datasets for transfer learning and pre-training tasks are yet to be constructed. Moreover, information mining from users needs to be further evaluated.
作者
吴诗苑
董庆兴
宋志君
张斌
Wu Shiyuan;Dong Qingxing;Song Zhijun;Zhang Bin(School of Information Management,Central China Normal University,Wuhan 430079;School of Journalism and Communication,Wuhan University,Wuhan 430072;Big Data Institute,Wuhan University,Wuhan 430072;School of Information Management,Nanjing University,Nanjing 210023)
出处
《情报学报》
CSSCI
CSCD
北大核心
2022年第6期651-661,共11页
Journal of the China Society for Scientific and Technical Information
基金
国家自然科学基金面上项目“面向群智感知大数据的群体评价模型与方法研究”(71871102),“基于注意力机制的学术信息动态推荐研究”(72074109)。
关键词
社交媒体
错误信息
深度学习
特征融合
自动检测
social media
misinformation
deep learning
feature fusion
automatic detection
