Stance Prediction for Contemporary Issues: Data and Experiments

Marjan Hosseinia, Eduard Dragut, Arjun Mukherjee


Abstract
We investigate whether pre-trained bidirectional transformers with sentiment and emotion information improve stance detection in long discussions of contemporary issues. As a part of this work, we create a novel stance detection dataset covering 419 different controversial issues and their related pros and cons collected by procon.org in nonpartisan format. Experimental results show that a shallow recurrent neural network with sentiment or emotion information can reach competitive results compared to fine-tuned BERT with 20x fewer parameters. We also use a simple approach that explains which input phrases contribute to stance detection.
Anthology ID:
2020.socialnlp-1.5
Volume:
Proceedings of the Eighth International Workshop on Natural Language Processing for Social Media
Month:
July
Year:
2020
Address:
Online
Venue:
SocialNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
32–40
Language:
URL:
https://aclanthology.org/2020.socialnlp-1.5
DOI:
10.18653/v1/2020.socialnlp-1.5
Bibkey:
Cite (ACL):
Marjan Hosseinia, Eduard Dragut, and Arjun Mukherjee. 2020. Stance Prediction for Contemporary Issues: Data and Experiments. In Proceedings of the Eighth International Workshop on Natural Language Processing for Social Media, pages 32–40, Online. Association for Computational Linguistics.
Cite (Informal):
Stance Prediction for Contemporary Issues: Data and Experiments (Hosseinia et al., SocialNLP 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.socialnlp-1.5.pdf
Dataset:
 2020.socialnlp-1.5.Dataset.zip
Video:
 http://slideslive.com/38929905
Code
 marjanhs/procon20
Data
Procon20