# Harvey Mudd College at SemEval-2019 Task 4: The Clint Buchanan Hyperpartisan News Detector

## TL;DR

This paper evaluates BERT-based models for hyperpartisan news detection, demonstrating high accuracy and robustness, with insights into the importance of local context and model consistency.

## Contribution

The study applies BERT to hyperpartisan news detection, optimizing parameters and analyzing the impact of word piece order on model performance.

## Key findings

- Best BERT model achieved 85% validation accuracy
- Model labeling is consistent across article slices
- Local context from word groups is crucial for accuracy

## Abstract

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.