Mapping the Media Landscape: Predicting Factual Reporting and Political Bias Through Web Interactions

TL;DR

This paper introduces a novel approach to assess news source bias by modeling web interactions over time, improving classification of factual reporting and political bias, validated on large datasets and benchmark challenges.

Contribution

It extends existing reliability estimation methods with reinforcement learning strategies to analyze hyperlink graphs, providing new insights into media bias detection.

Findings

Significant performance improvement in bias classification at media source level

Outperforms previous results on CLEF 2023 CheckThat! challenge

Releases the largest annotated dataset of news sources with bias labels

Abstract

Bias assessment of news sources is paramount for professionals, organizations, and researchers who rely on truthful evidence for information gathering and reporting. While certain bias indicators are discernible from content analysis, descriptors like political bias and fake news pose greater challenges. In this paper, we propose an extension to a recently presented news media reliability estimation method that focuses on modeling outlets and their longitudinal web interactions. Concretely, we assess the classification performance of four reinforcement learning strategies on a large news media hyperlink graph. Our experiments, targeting two challenging bias descriptors, factual reporting and political bias, showed a significant performance improvement at the source media level. Additionally, we validate our methods on the CLEF 2023 CheckThat! Lab challenge, outperforming the reported results in both, F1-score and the official MAE metric. Furthermore, we contribute by releasing the largest annotated dataset of news source media, categorized with factual reporting and political bias labels. Our findings suggest that profiling news media sources based on their hyperlink interactions over time is feasible, offering a bird's-eye view of evolving media landscapes.