We Can Detect Your Bias: Predicting the Political Ideology of News Articles

TL;DR

This paper presents a new dataset and methods for predicting the political bias of news articles, achieving significant improvements over existing models by addressing source bias and using adversarial adaptation.

Contribution

It introduces a large, balanced dataset of news articles annotated for political bias and proposes novel adversarial and triplet loss techniques for better prediction accuracy.

Findings

Model outperforms state-of-the-art Transformers in bias prediction.

Background source information improves prediction accuracy.

Adversarial media adaptation enhances model robustness.

Abstract

We explore the task of predicting the leading political ideology or bias of news articles. First, we collect and release a large dataset of 34,737 articles that were manually annotated for political ideology -left, center, or right-, which is well-balanced across both topics and media. We further use a challenging experimental setup where the test examples come from media that were not seen during training, which prevents the model from learning to detect the source of the target news article instead of predicting its political ideology. From a modeling perspective, we propose an adversarial media adaptation, as well as a specially adapted triplet loss. We further add background information about the source, and we show that it is quite helpful for improving article-level prediction. Our experimental results show very sizable improvements over using state-of-the-art pre-trained Transformers in this challenging setup.