Cost-Sensitive BERT for Generalisable Sentence Classification with Imbalanced Data
This addresses propaganda detection, a domain-specific task with imbalanced and evolving news data, offering an incremental improvement for NLP applications in misinformation analysis.
The paper tackled the problem of BERT's poor generalization in sentence classification when training and test data are dissimilar, especially with imbalanced classes, by introducing a statistical similarity measure and cost-weighting method, achieving the second-highest score on the Propaganda Techniques Corpus.
The automatic identification of propaganda has gained significance in recent years due to technological and social changes in the way news is generated and consumed. That this task can be addressed effectively using BERT, a powerful new architecture which can be fine-tuned for text classification tasks, is not surprising. However, propaganda detection, like other tasks that deal with news documents and other forms of decontextualized social communication (e.g. sentiment analysis), inherently deals with data whose categories are simultaneously imbalanced and dissimilar. We show that BERT, while capable of handling imbalanced classes with no additional data augmentation, does not generalise well when the training and test data are sufficiently dissimilar (as is often the case with news sources, whose topics evolve over time). We show how to address this problem by providing a statistical measure of similarity between datasets and a method of incorporating cost-weighting into BERT when the training and test sets are dissimilar. We test these methods on the Propaganda Techniques Corpus (PTC) and achieve the second-highest score on sentence-level propaganda classification.