Agreement Prediction of Arguments in Cyber Argumentation for Detecting Stance Polarity and Intensity

Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 5746–5758 July 5 - 10, 2020. c⃝2020 Association for Computational Linguistics 5746 Agreement Prediction of Arguments in Cyber Argumentation for Detecting Stance Polarity and Intensity Joseph W Sirrianni, Xiaoqing “Frank” Liu, Douglas Adams University of Arkansas, Fayetteville, AR, USA. {jwsirria,frankliu,djadams}@uark.edu Abstract In online debates, users express different levels of agreement/disagreement with one another’s arguments and ideas. Often levels of agree- ment/disagreement are implicit in the text and must be predicted to analyze collective opin- ions. Existing stance detection methods pre- dict the polarity of a post’s stance toward a topic or post, but don’t consider the stance’s degree of intensity. We introduce a new re- search problem, stance polarity and intensity prediction in response relationships between posts. This problem is challenging because differences in stance intensity are often sub- tle and require nuanced language understand- ing. Cyber argumentation research has shown that incorporating both stance polarity and in- tensity data in online debates leads to better discussion analysis. We explore five different learning models: Ridge-M regression, Ridge- S regression, SVR-RF-R, pkudblab-PIP, and T-PAN-PIP for predicting stance polarity and intensity in argumentation. These models are evaluated using a new dataset for stance polar- ity and intensity prediction collected using a cyber argumentation platform. The SVR-RF- R model performs best for prediction of stance polarity with an accuracy of 70.43% and inten- sity with RMSE of 0.596. This work is the first to train models for predicting a post’s stance polarity and intensity in one combined value in cyber argumentation with reasonably good accuracy. 1 Introduction Many major online and social media and network- ing sites, such as Facebook, Twitter, and Wikipedia, have taken over as the new public forum for people to discuss and debate issues of national and interna- tional importance. With more participants in these debates than ever before, the volume of unstruc- tured discourse data continues to increase, and the need for automatic processing of this data is preva- lent. A critical task in processing online debates is to automatically determine the different argu- mentative relationships between online posts in a discussion. These relationships typically consist of a stance polarity (i.e., whether a post is supporting, opposing, or is neutral toward another post) and the degree of intensity of the stance. Automatically determining these types of re- lationships from a given text is a goal in both stance detection and argumentation mining re- search. Stance detection models seek to automati- cally determine a text’s stance polarity (Favoring, Opposing, or Neutral) toward another text or topic based on its textual information (Mohammad et al., 2016). Likewise, argumentation mining seeks to determine the stance relationship (Supporting, At- tacking, or Neutral) between argumentation compo- nents in a text (Stede and Schneider, 2018). How- ever, in both cases, attention is only paid to the stance’s polarity, while the intensity of the relation- ship is often ignored. Some studies have tried to incorporate intensity into their predictions by ex- panding …

Content truncated to 500 words