Better Reasoning Behind Classification Predictions with BERT for Fake News Detection
This addresses the need for better interpretability in fake news detection models, though it is incremental as it builds on existing methods like CAM and BERT.
The study tackled the problem of insufficient reasoning behind classification predictions in fake news detection by analyzing representation space quality and proposing a modified Class Activation Mapping (CAM) for interpretability, achieving robust performance with a naive BERT model topped with a learnable linear layer.
Fake news detection has become a major task to solve as there has been an increasing number of fake news on the internet in recent years. Although many classification models have been proposed based on statistical learning methods showing good results, reasoning behind the classification performances may not be enough. In the self-supervised learning studies, it has been highlighted that a quality of representation (embedding) space matters and directly affects a downstream task performance. In this study, a quality of the representation space is analyzed visually and analytically in terms of linear separability for different classes on a real and fake news dataset. To further add interpretability to a classification model, a modification of Class Activation Mapping (CAM) is proposed. The modified CAM provides a CAM score for each word token, where the CAM score on a word token denotes a level of focus on that word token to make the prediction. Finally, it is shown that the naive BERT model topped with a learnable linear layer is enough to achieve robust performance while being compatible with CAM.