Deep Breath：一个用于解决在线错误信息的机器学习浏览器扩展
Deep Breath: A Machine Learning Browser Extension to Tackle Online Misinformation
摘要：
在过去的十年里，媒体格局发生了根本性的变化。随着越来越多的公众通过在线资源了解时事，竞争也随着媒体争夺关注而加剧。这种竞争促使一些网络媒体发布耸人听闻的内容来吸引读者的注意力。这种做法可能会歪曲事实，误导读者对事件性质的认识，从而威胁到民主。本文提出了一种新的系统，用于检测，处理和警告用户有关误导性内容在线打击所造成的威胁的错误信息。通过在32，000个点击诱饵新闻文章标题的现有数据集上训练机器学习模型，该模型预测标题的耸人听闻程度，然后与Web浏览器扩展接口，该扩展基于现有的设计原则构建独特的内容警告通知并合并模型的预测。这项研究为机器学习和以人为本的安全性做出了新的贡献，并为未来的研究提供了有希望的发现。通过警告用户可能正在查看错误信息，可以防止自发反应，帮助用户深呼吸并以清晰的头脑接近在线媒体。
英文摘要：
Over the past decade, the media landscape has seen a radical shift. As more of the public stay informed of current events via online sources, competition has grown as outlets vie for attention. This competition has prompted some online outlets to publish sensationalist and alarmist content to grab readers' attention. Such practices may threaten democracy by distorting the truth and misleading readers about the nature of events. This paper proposes a novel system for detecting, processing, and warning users about misleading content online to combat the threats posed by misinformation. By training a machine learning model on an existing dataset of 32,000 clickbait news article headlines, the model predicts how sensationalist a headline is and then interfaces with a web browser extension which constructs a unique content warning notification based on existing design principles and incorporates the models' prediction. This research makes a novel contribution to machine learning and human-centred security with promising findings for future research. By warning users when they may be viewing misinformation, it is possible to prevent spontaneous reactions, helping users to take a deep breath and approach online media with a clear mind.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
