ToxBuster: In-game Chat Toxicity Buster with BERT
Preprint
Abstract
Detecting toxicity in online spaces is challenging and an ever more pressing problem given the increase in social media and gaming consumption. We introduce ToxBuster, a simple and scalable model trained on a relatively large dataset of 194k lines of game chat from Rainbow Six Siege and For Honor, carefully annotated for different kinds of toxicity. Compared to the existing state-of-the-art, ToxBuster achieves 82.95% (+7) in precision and 83.56% (+57) in recall. This improvement is obtained by leveraging past chat history and metadata. We also study the implication towards real-time and post-game moderation as well as the model transferability from one game to another.Keywords
All Related Versions
This publication has 19 references indexed in Scilit:
- PROVOKE: Toxicity trigger detection in conversations from the top 100 subredditsData and Information Management, 2022
- An Empirical Survey of the Effectiveness of Debiasing Techniques for Pre-trained Language Models2022
- Improving Contextual Language Models for Response Retrieval in Multi-Turn Conversation2020
- Measuring and Characterizing Hate Speech on News Websites2020
- Toxicity Detection: Does Context Really Matter?2020
- Language (Technology) is Power: A Critical Survey of “Bias” in NLP2020
- On Measuring Social Biases in Sentence Encoders2019
- Hate Speech on Twitter: A Pragmatic Approach to Collect Hateful and Offensive Expressions and Perform Hate Speech DetectionIEEE Access, 2018
- The Impact of Toxic Language on the Health of Reddit Communities2017
- Can a video game company tame toxic behaviour?Nature, 2016