A Stylometric Inquiry into Hyperpartisan and Fake News
Martin Potthast Johannes Kiesel Kevin Reinartz Janek Bevendorff Benno Stein Bauhaus-Universit t Weimar <<first name>>.<<last
Abstract
This paper reports on a writing style analysis of hyperpartisan (i.e., extremely one-sided) news in connection to fake news. It presents a large corpus of 1,627 articles that were manually fact-checked by professional journalists from BuzzFeed. The articles originated from 9 well-known political publishers, 3 each from the mainstream, the hyperpartisan left-wing, and the hyperpartisan right-wing. In sum, the corpus contains 299 fake news, 97% of which originated from hyperpartisan publishers.
中文速览
假新闻(fake news)在社交媒体上的泛滥令人忧虑,但靠逐条核实事实来自动检测它既费时又困难,于是研究者转而问:假新闻在写作风格上有没有规律可循?本文基于一个由BuzzFeed专业记者人工核查的1627篇新闻语料库,首次系统分析了极端党派性新闻(hyperpartisan news)与假新闻在文体风格上的关联,并引入"去伪装"(Unmasking)这一元学习方法来衡量不同类别文本之间的风格相似度。研究发现,极端左翼与极端右翼新闻在写作风格上彼此高度相似,却都与主流新闻风格迥异,基于风格的分类器能较好地区分党派性新闻与主流新闻(F₁=0.78),也能有效识别讽刺性文章(F₁=0.81),但单凭风格无法可靠地检测假新闻本身(F₁=0.46)。这一结论意义重大:虽然风格特征不足以直接判定真假,但可以用来对新闻进行党派性预筛查,从而为后续事实核查系统大幅缩小目标范围、提升效率。
原文 arXiv:1702.05638;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1702.05638v1