ur-iw-hnt at GermEval 2021: An Ensembling Strategy with Multiple BERT Models

Hoai Nam Tran, Udo Kruschwitz


Abstract
This paper describes our approach (ur-iw-hnt) for the Shared Task of GermEval2021 to identify toxic, engaging, and fact-claiming comments. We submitted three runs using an ensembling strategy by majority (hard) voting with multiple different BERT models of three different types: German-based, Twitter-based, and multilingual models. All ensemble models outperform single models, while BERTweet is the winner of all individual models in every subtask. Twitter-based models perform better than GermanBERT models, and multilingual models perform worse but by a small margin.
Anthology ID:
2021.germeval-1.12
Volume:
Proceedings of the GermEval 2021 Shared Task on the Identification of Toxic, Engaging, and Fact-Claiming Comments
Month:
September
Year:
2021
Address:
Duesseldorf, Germany
Venue:
GermEval
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
83–87
Language:
URL:
https://aclanthology.org/2021.germeval-1.12
DOI:
Bibkey:
Cite (ACL):
Hoai Nam Tran and Udo Kruschwitz. 2021. ur-iw-hnt at GermEval 2021: An Ensembling Strategy with Multiple BERT Models. In Proceedings of the GermEval 2021 Shared Task on the Identification of Toxic, Engaging, and Fact-Claiming Comments, pages 83–87, Duesseldorf, Germany. Association for Computational Linguistics.
Cite (Informal):
ur-iw-hnt at GermEval 2021: An Ensembling Strategy with Multiple BERT Models (Tran & Kruschwitz, GermEval 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.germeval-1.12.pdf
Code
 hn-tran/germeval2021