Revisiting Rumour Stance Classification: Dealing with Imbalanced Data

Yue Li, Carolina Scarton


Abstract
Correctly classifying stances of replies can be significantly helpful for the automatic detection and classification of online rumours. One major challenge is that there are considerably more non-relevant replies (comments) than informative ones (supports and denies), making the task highly imbalanced. In this paper we revisit the task of rumour stance classification, aiming to improve the performance over the informative minority classes. We experiment with traditional methods for imbalanced data treatment with feature- and BERT-based classifiers. Our models outperform all systems in RumourEval 2017 shared task and rank second in RumourEval 2019.
Anthology ID:
2020.rdsm-1.4
Volume:
Proceedings of the 3rd International Workshop on Rumours and Deception in Social Media (RDSM)
Month:
December
Year:
2020
Address:
Barcelona, Spain (Online)
Venue:
RDSM
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
38–44
Language:
URL:
https://aclanthology.org/2020.rdsm-1.4
DOI:
Bibkey:
Cite (ACL):
Yue Li and Carolina Scarton. 2020. Revisiting Rumour Stance Classification: Dealing with Imbalanced Data. In Proceedings of the 3rd International Workshop on Rumours and Deception in Social Media (RDSM), pages 38–44, Barcelona, Spain (Online). Association for Computational Linguistics.
Cite (Informal):
Revisiting Rumour Stance Classification: Dealing with Imbalanced Data (Li & Scarton, RDSM 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.rdsm-1.4.pdf