Team yeon-zi at SemEval-2019 Task 4: Hyperpartisan News Detection by De-noising Weakly-labeled Data

This paper describes our system submitted to SemEval-2019 Task 4: Hyperpartisan News Detection. We focus on removing the inherent noise in the hyperpartisanship dataset from both data-level and model-level by leveraging semi-supervised pseudo-labels and the stateof-the-art BERT model. Our model achieves 75.8% accuracy in the final by-article dataset without ensemble learning.