Abstract
This work presents a systematic search of various model architecture configurations and data cleaning methods. The study evaluates the impact of data cleaning methods on the obtained results. Additionally, we demonstrate that a combination of CNN and Encoder-only models such as BERTweet outperforms FNNs. Moreover, by utilizing data augmentation, we are able to overcome the challenge of data imbalance.- Anthology ID:
- 2024.case-1.24
- Volume:
- Proceedings of the 7th Workshop on Challenges and Applications of Automated Extraction of Socio-political Events from Text (CASE 2024)
- Month:
- March
- Year:
- 2024
- Address:
- St. Julians, Malta
- Editors:
- Ali Hürriyetoğlu, Hristo Tanev, Surendrabikram Thapa, Gökçe Uludoğan
- Venues:
- CASE | WS
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 178–184
- Language:
- URL:
- https://aclanthology.org/2024.case-1.24
- DOI:
- Cite (ACL):
- Ghazaleh Mahmoudi and Sauleh Eetemadi. 2024. IUST at ClimateActivism 2024: Towards Optimal Stance Detection: A Systematic Study of Architectural Choices and Data Cleaning Techniques. In Proceedings of the 7th Workshop on Challenges and Applications of Automated Extraction of Socio-political Events from Text (CASE 2024), pages 178–184, St. Julians, Malta. Association for Computational Linguistics.
- Cite (Informal):
- IUST at ClimateActivism 2024: Towards Optimal Stance Detection: A Systematic Study of Architectural Choices and Data Cleaning Techniques (Mahmoudi & Eetemadi, CASE-WS 2024)
- PDF:
- https://preview.aclanthology.org/ingest-2024-clasp/2024.case-1.24.pdf