Disentangling Structure and Style: Political Bias Detection in News by Inducing Document Hierarchy
This work addresses the issue of limited generalizability in political bias detection for news analysis, though it appears incremental by building on existing hierarchical attention models.
The paper tackles the problem of detecting political bias in news articles by addressing overfitting to writing styles, resulting in a more robust and style-agnostic approach that outperforms previous methods in robustness and accuracy.
We address an important gap in detecting political bias in news articles. Previous works that perform document classification can be influenced by the writing style of each news outlet, leading to overfitting and limited generalizability. Our approach overcomes this limitation by considering both the sentence-level semantics and the document-level rhetorical structure, resulting in a more robust and style-agnostic approach to detecting political bias in news articles. We introduce a novel multi-head hierarchical attention model that effectively encodes the structure of long documents through a diverse ensemble of attention heads. While journalism follows a formalized rhetorical structure, the writing style may vary by news outlet. We demonstrate that our method overcomes this domain dependency and outperforms previous approaches for robustness and accuracy. Further analysis and human evaluation demonstrate the ability of our model to capture common discourse structures in journalism. Our code is available at: https://github.com/xfactlab/emnlp2023-Document-Hierarchy