Recognizing stances in online debates

Abstract

This paper presents an unsupervised opinion analysis method for debate-side classification, i.e., recognizing which stance a person is taking in an online debate. In order to handle the complexities of this genre, we mine the web to learn associations that are indicative of opinion stances in debates. We combine this knowledge with discourse information, and formulate the debate side classification task as an Integer Linear Programming problem. Our results show that our method is substantially better than challenging baseline methods.

Automatic Summarization of Online Debates

“…Somasundaran and Wiebe (2009) noticed that in online debate posts, people debate issues, express their favorites, oppose other stances, and argue why their thoughts are correct. To determine positive sentiment about one target, expressing negative sentiment about the other side is a key target.…”

Annotating Speaker Stance in Discourse: The Brexit Blog Corpus

“…The first studies on stance detection and classification from a computational perspective used data derived from ideological online debates (Somasundaran and Wiebe 2010). They created a lexicon with positive and negative entries, and showed that the sentiment-and argument-based systems outperform the baseline ones in overall accuracy (63.93 %).…”

Proceedings of the 2nd Workshop on Argumentation Mining

“…Supervised learning has been used in almost all of the current approaches for stance classification, in which a large set of data has been collected and annotated in order to be used as training data for classifiers. In (Somasundaran and Wiebe, 2010), a lexicon for detecting argument trigger expressions was created and subsequently leveraged to identify arguments.…”

“…A related older strand of research (that uses the term 'argumentative structure' in a related but different sense than ours) has investigated automatically classifying the sentences of a scientific article's abstract or full text in terms of their contribution of new knowledge to a field (e.g., Liakata et al 2012, Teufel 2010, Mizuta et al 2005). In addition, argumentation mining has ties to sentiment analysis (e.g., Somasundaran and Wiebe 2010). To date there are few corpora with annotations for argumentation mining research ) although corpora with annotations for argument sub-components have recently become available (e.g., .…”

“…Stance has been defined as the overall position toward an idea, object or proposition (Somasundaran and Wiebe, 2010). There has been growing interest in stance classification particularly for online debates (Walker et al, 2012a), (Hasan and Ng, 2013).…”