UCCNLP@SMM4H’22:Label distribution aware long-tailed learning with post-hoc posterior calibration applied to text classification

Trust, Paul; Kadusabe, Provia; Zahran, Ahmed; Minghim, Rosane; Omala, Kizito

UCCNLP@SMM4H’22:Label distribution aware long-tailed learning with post-hoc posterior calibration applied to text classification

cb

Files

2022.smm4h-1.26.pdf(230.18 KB)

Published Version

Date

2022-10

Authors

Trust, Paul

Kadusabe, Provia

Zahran, Ahmed

Minghim, Rosane

Omala, Kizito

Publisher

Association for Computational Linguistics

Abstract

The paper describes our submissions for the Social Media Mining for Health (SMM4H) workshop 2022 shared tasks. We participated in 2 tasks: (1) classification of adverse drug events (ADE) mentions in english tweets (Task-1a) and (2) classification of self-reported intimate partner violence (IPV) on twitter (Task 7). We proposed an approach that uses RoBERTa (A Robustly Optimized BERT Pretraining Approach) fine-tuned with a label distribution-aware margin loss function and post-hoc posterior calibration for robust inference against class imbalance. We achieved a 4% and 1 % increase in performance on IPV and ADE respectively when compared with the traditional fine-tuning strategy with unweighted cross-entropy loss.

Keywords

Social Media Mining for Health

Citation

Trust, P., Kadusabe, P., Zahran, A., Minghim, R. and Omala, K. (2022) 'UCCNLP@SMM4H’22:Label distribution aware long-tailed learning with post-hoc posterior calibration applied to text classification', Proceedings of The Seventh Workshop on Social Media Mining for Health Applications, Workshop & Shared Task, Gyeongju, Republic of Korea, 12-17 October, pp. 90-94. Available at: https://aclanthology.org/2022.smm4h-1.26.pdf (Accessed: 9 November 2022)