Skip to main content
QUICK REVIEW

[论文解读] Beyond News Contents: The Role of Social Context for Fake News Detection

Kai Shu, Suhang Wang|arXiv (Cornell University)|Dec 20, 2017
Misinformation and Its Impacts参考文献 34被引用 106
一句话总结

TriFN 利用出版商-新闻关系与用户互动,在 FakeNewsNet(BuzzFeed 和 PolitiFact)上检测假新闻,显著优于基线方法。

ABSTRACT

Social media is becoming popular for news consumption due to its fast dissemination, easy access, and low cost. However, it also enables the wide propagation of fake news, i.e., news with intentionally false information. Detecting fake news is an important task, which not only ensures users to receive authentic information but also help maintain a trustworthy news ecosystem. The majority of existing detection algorithms focus on finding clues from news contents, which are generally not effective because fake news is often intentionally written to mislead users by mimicking true news. Therefore, we need to explore auxiliary information to improve detection. The social context during news dissemination process on social media forms the inherent tri-relationship, the relationship among publishers, news pieces, and users, which has potential to improve fake news detection. For example, partisan-biased publishers are more likely to publish fake news, and low-credible users are more likely to share fake news. In this paper, we study the novel problem of exploiting social context for fake news detection. We propose a tri-relationship embedding framework TriFN, which models publisher-news relations and user-news interactions simultaneously for fake news classification. We conduct experiments on two real-world datasets, which demonstrate that the proposed approach significantly outperforms other baseline methods for fake news detection.

研究动机与目标

  • 在社交媒体时代,单靠内容不足以判断,需要推动假新闻检测。
  • 定义一个三元关系(出版商、新闻、用户)以建模传播动态。
  • 开发 TriFN,以从内容、社交上下文和互动中学习联合表征。
  • 通过真实世界数据集实验来证明有效性。

提出的方法

  • 提出 TriFN,一种三元关系嵌入框架,结合五个组件:新闻内容嵌入、用户嵌入、用户-新闻交互嵌入、出版商-新闻关系嵌入,以及半监督分类。
  • 使用非负矩阵分解(NMF)分别对新闻内容和用户进行嵌入(D,V)和(U,T)。
  • 引入用户可信度分数,通过将用户潜在特征与真/假标签相连的权重距离项来引导用户-新闻交互学习。
  • 通过对带标签的出版商使用出版商-新闻关系嵌入来对出版商的党派偏见进行建模,使出版商特征与党派分数对齐。
  • 实现一个半监督分类器,采用从新闻潜在特征到标签的线性映射,通过在 D,U,V,T,p,q 上交替最小二乘法进行优化。

实验结果

研究问题

  • RQ1TriFN 是否能够通过联合建模出版商党派偏见和用户参与来改进假新闻分类?
  • RQ2出版商偏见建模和用户参与学习如何提升 TriFN 的性能?
  • RQ3在用户参与有限时,TriFN 是否能够实现早期假新闻检测?

主要发现

数据集方法准确率精确度召回率F1
BuzzFeedTriFN0.864±0.0260.849±0.0400.893±0.0130.870±0.019
PolitiFactTriFN0.878±0.0170.867±0.0340.893±0.0230.880±0.015
  • TriFN 在两个真实世界数据集(BuzzFeed 和 PolitiFact)的 F1 分数及其他指标上显著优于基线方法。
  • 在 BuzzFeed 上,TriFN 的 F1 为 0.870,优于 RST(0.633)和 LIWC+Castillo(0.822)。
  • 在 PolitiFact 上,TriFN 的 F1 为 0.880,优于 RST(0.615)和 LIWC+Castillo(0.843)。
  • 该框架利用出版商党派偏见和用户可信度来改进新闻表征和假新闻预测。
  • 在 FakeNewsNet 上进行实验,采用 80-20 的训练测试划分,结果在 10 次运行上取平均。

更好的研究,从现在开始

从论文设计到论文写作,大幅缩短您的研究时间。

无需绑定信用卡

本解读由 AI 生成,并经人工编辑审核。