This paper describes the UM-IU@LING's system for the SemEval 2019 Task 6: OffensEval. We take a mixed approach to identify and categorize hate speech in social media. In subtask A, we fine-tuned a BERT based classifier to detect abusive content in tweets, achieving a macro F1 score of 0.8136 on the test data, thus reaching the 3rd rank out of 103 submissions. In subtasks B and C, we used a linear SVM with selected character n-gram features. For subtask C, our system could identify the target of abuse with a macro F1 score of 0.5243, ranking it 27th out of 65 submissions.
@article{arxiv.1904.03450,
title = {UM-IU@LING at SemEval-2019 Task 6: Identifying Offensive Tweets Using BERT and SVMs},
author = {Jian Zhu and Zuoyu Tian and Sandra Kübler},
journal= {arXiv preprint arXiv:1904.03450},
year = {2019}
}