Abstract
In this work, we provide insight into three key aspects related to predicting argument convincingness. First, we explicitly display the power that text length possesses for predicting convincingness in an unsupervised setting. Second, we show that a bag-of-words embedding model posts state-of-the-art on a dataset of arguments annotated for convincingness, outperforming an SVM with numerous hand-crafted features as well as recurrent neural network models that attempt to capture semantic composition. Finally, we assess the feasibility of integrating external knowledge when predicting convincingness, as arguments are often more convincing when they contain abundant information and facts. We finish by analyzing the correlations between the various models we propose.- Anthology ID:
- I17-1035
- Volume:
- Proceedings of the Eighth International Joint Conference on Natural Language Processing (Volume 1: Long Papers)
- Month:
- November
- Year:
- 2017
- Address:
- Taipei, Taiwan
- Venue:
- IJCNLP
- SIG:
- Publisher:
- Asian Federation of Natural Language Processing
- Note:
- Pages:
- 342–351
- Language:
- URL:
- https://aclanthology.org/I17-1035
- DOI:
- Cite (ACL):
- Peter Potash, Robin Bhattacharya, and Anna Rumshisky. 2017. Length, Interchangeability, and External Knowledge: Observations from Predicting Argument Convincingness. In Proceedings of the Eighth International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 342–351, Taipei, Taiwan. Asian Federation of Natural Language Processing.
- Cite (Informal):
- Length, Interchangeability, and External Knowledge: Observations from Predicting Argument Convincingness (Potash et al., IJCNLP 2017)
- PDF:
- https://preview.aclanthology.org/auto-file-uploads/I17-1035.pdf