Harvey Mudd College at SemEval-2019 Task 4: The Clint Buchanan Hyperpartisan News Detector
This work addresses the problem of detecting biased news for media analysis, but it is incremental as it applies an existing method to a specific dataset.
The paper tackled hyperpartisan news detection by applying BERT models, achieving 85% accuracy on a validation set and 77% on a test set, and found that the model relies heavily on local context as shuffling word pieces reduced accuracy.
We investigate the recently developed Bidirectional Encoder Representations from Transformers (BERT) model for the hyperpartisan news detection task. Using a subset of hand-labeled articles from SemEval as a validation set, we test the performance of different parameters for BERT models. We find that accuracy from two different BERT models using different proportions of the articles is consistently high, with our best-performing model on the validation set achieving 85% accuracy and the best-performing model on the test set achieving 77%. We further determined that our model exhibits strong consistency, labeling independent slices of the same article identically. Finally, we find that randomizing the order of word pieces dramatically reduces validation accuracy (to approximately 60%), but that shuffling groups of four or more word pieces maintains an accuracy of about 80%, indicating the model mainly gains value from local context.