Understanding Politics via Contextualized Discourse Processing

Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 1353–1367 November 7–11, 2021. c⃝2021 Association for Computational Linguistics 1353 Understanding Politics via Contextualized Discourse Processing Rajkumar Pujari and Dan Goldwasser Department of Computer Science Purdue University {rpujari,dgoldwas}@purdue.edu Abstract Politicians often have underlying agendas when reacting to events. Arguments in con- texts of various events reflect a fairly consis- tent set of agendas for a given entity. In spite of recent advances in Pretrained Language Mod- els, those text representations are not designed to capture such nuanced patterns. In this paper, we propose a Compositional Reader model consisting of encoder and composer modules, that captures and leverages such information to generate more effective representations for entities, issues, and events. These represen- tations are contextualized by tweets, press re- leases, issues, news articles, and participating entities. Our model processes several docu- ments at once and generates composed repre- sentations for multiple entities over several is- sues or events. Via qualitative and quantitative empirical analysis, we show that these repre- sentations are meaningful and effective. 1 Introduction Over the last decade political discourse has moved from traditional outlets to social media. This pro- cess, starting in the ’08 U.S. presidential elections, has peaked in recent years, with former-president Trump announcing the firing of top officials as well as policy decisions over Twitter. This presents a new challenge to the NLP community, how can this massive amount of political content be used to create principled representations of politicians, their stances on issues and legislative preferences? This is not an easy challenge as in political texts perspective is often subtle rather than explicit (Fan et al., 2019). Choices of mentioning or omitting certain entities or attributes can reveal the author’s agenda. For example, tweeting “mass shootings are due to a huge mental health problem” in reac- tion to a mass shooting is likely to be indicative of opposing gun control measures, despite the lack of an explicit stance in the text. Recent advances in Pretrained Language Models (PLMs) in NLP (Devlin et al., 2019; Yang et al., 2019; Liu et al., 2019) have greatly improved word representations via contextualized embeddings and powerful transformer units, however such repre- sentations alone are not enough to capture nuanced biases in political discourse. Two of the key reasons are: (i) they do not directly focus on entity/issue- centric data and (ii) they only represent linguistic context rather external political context. Our main insight is that effectively detecting such bias from text requires modeling the broader political context of the document. This can include understanding relevant facts related to the event addressed in the text, the ideological leanings and perspectives expressed by the author in the past, and the sentiment/attitude of the author towards the entities referenced in the text. We suggest that this holistic view can be obtained by combining information from multiple sources, which can be of varying types, such as news articles, social media posts, quotes from press releases and historical beliefs expressed …

Content truncated to 500 words