Predicting Sentence-Level Factuality of News and Bias of Media Outlets

TL;DR

This paper introduces FactNews, a large annotated dataset for sentence-level factuality and media bias, and demonstrates methods to predict news reliability and bias, including for Portuguese news sources.

Contribution

The creation of the FactNews dataset with expert annotations and the development of classification methods for factuality and bias prediction, including for Portuguese news.

Findings

Biased sentences tend to have more words and emotional content.

Fine-grained analysis improves reliability prediction.

Baseline models perform effectively on Portuguese news.

Abstract

Automated news credibility and fact-checking at scale require accurately predicting news factuality and media bias. This paper introduces a large sentence-level dataset, titled "FactNews", composed of 6,191 sentences expertly annotated according to factuality and media bias definitions proposed by AllSides. We use FactNews to assess the overall reliability of news sources, by formulating two text classification problems for predicting sentence-level factuality of news reporting and bias of media outlets. Our experiments demonstrate that biased sentences present a higher number of words compared to factual sentences, besides having a predominance of emotions. Hence, the fine-grained analysis of subjectivity and impartiality of news articles provided promising results for predicting the reliability of media outlets. Finally, due to the severity of fake news and political polarization in Brazil, and the lack of research for Portuguese, both dataset and baseline were proposed for Brazilian Portuguese.