Abstract
We present our results for the shared tasks 2, 4 and 9 at the SMM4H Workshop at COLING 2022 achieved by succesfully fine-tuning pre-trained language models to the downstream tasks. We identify the occurence of code-switching in the test data for task 2 as a possible source of considerable performance degradation on the test set scores. We successfully exploit structural linguistic similarities in the datasets of tasks 4 and 9 for training on joined datasets, scoring first in task 9 and on par with SOTA in task 4.- Anthology ID:
- 2022.smm4h-1.29
- Volume:
- Proceedings of The Seventh Workshop on Social Media Mining for Health Applications, Workshop & Shared Task
- Month:
- October
- Year:
- 2022
- Address:
- Gyeongju, Republic of Korea
- Editors:
- Graciela Gonzalez-Hernandez, Davy Weissenbacher
- Venue:
- SMM4H
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 103–107
- Language:
- URL:
- https://aclanthology.org/2022.smm4h-1.29
- DOI:
- Cite (ACL):
- Daniel Claeser and Samantha Kent. 2022. Fraunhofer FKIE @ SMM4H 2022: System Description for Shared Tasks 2, 4 and 9. In Proceedings of The Seventh Workshop on Social Media Mining for Health Applications, Workshop & Shared Task, pages 103–107, Gyeongju, Republic of Korea. Association for Computational Linguistics.
- Cite (Informal):
- Fraunhofer FKIE @ SMM4H 2022: System Description for Shared Tasks 2, 4 and 9 (Claeser & Kent, SMM4H 2022)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-3/2022.smm4h-1.29.pdf