We Can Detect Your Bias: Predicting the Political Ideology of News Articles
This work addresses the challenge of detecting political bias in news articles, which is important for media analysis and fact-checking, though it is incremental as it builds on existing methods with new adaptations.
The paper tackles the problem of predicting the political ideology (left, center, or right) of news articles by introducing a manually annotated dataset of 34,737 articles and a challenging experimental setup where test media are unseen during training, resulting in very sizable improvements over state-of-the-art pre-trained Transformers.
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.