Detection of Stance-Related Characteristics in Social Media Text

Abstract

In this paper, we present a study for the identification of stancerelated features in text data from social media. Based on our previous work on stance and our findings on stance patterns, we detected stance-related characteristics in a data set from Twitter and Facebook. We extracted various corpus-, quantitative- and computational-based features that proved to be significant for six stance categories (contrariety, hypotheticality, necessity, prediction, source of knowledge, and uncertainty), and we tested them in our data set. The results of a preliminary clustering method are presented and discussed as a starting point for future contributions in the field. The results of our experiments showed a strong correlation between different characteristics and stance constructions, which can lead us to a methodology for automatic stance annotation of these data.