Mama Edha at SemEval-2017 Task 8: Stance Classification with CNN and Rules

Marianela García Lozano, Hanna Lilja, Edward Tjörnhammar, Maja Karasalo


Abstract
For the competition SemEval-2017 we investigated the possibility of performing stance classification (support, deny, query or comment) for messages in Twitter conversation threads related to rumours. Stance classification is interesting since it can provide a basis for rumour veracity assessment. Our ensemble classification approach of combining convolutional neural networks with both automatic rule mining and manually written rules achieved a final accuracy of 74.9% on the competition’s test data set for Task 8A. To improve classification we also experimented with data relabeling and using the grammatical structure of the tweet contents for classification.
Anthology ID:
S17-2084
Volume:
Proceedings of the 11th International Workshop on Semantic Evaluation (SemEval-2017)
Month:
August
Year:
2017
Address:
Vancouver, Canada
Editors:
Steven Bethard, Marine Carpuat, Marianna Apidianaki, Saif M. Mohammad, Daniel Cer, David Jurgens
Venue:
SemEval
SIG:
SIGLEX
Publisher:
Association for Computational Linguistics
Note:
Pages:
481–485
Language:
URL:
https://aclanthology.org/S17-2084
DOI:
10.18653/v1/S17-2084
Bibkey:
Cite (ACL):
Marianela García Lozano, Hanna Lilja, Edward Tjörnhammar, and Maja Karasalo. 2017. Mama Edha at SemEval-2017 Task 8: Stance Classification with CNN and Rules. In Proceedings of the 11th International Workshop on Semantic Evaluation (SemEval-2017), pages 481–485, Vancouver, Canada. Association for Computational Linguistics.
Cite (Informal):
Mama Edha at SemEval-2017 Task 8: Stance Classification with CNN and Rules (García Lozano et al., SemEval 2017)
Copy Citation:
PDF:
https://aclanthology.org/S17-2084.pdf