Who Sides with Whom? Towards Computational Construction of Discourse Networks for Political Debates

Abstract

Understanding the structures of political debates (which actors make what claims) is essential for understanding democratic political decision-making. The vision of computational construction of such discourse networks from newspaper reports brings together political science and natural language processing. This paper presents three contributions towards this goal: (a) a requirements analysis, linking the task to knowledge base population; (b) a first release of an annotated corpus of claims on the topic of migration, based on German newspaper reports; (c) initial modeling results.

An Environment for Relational Annotation of Political Debates

“…Particularly helpful for high recall was the integration of ML-based pseudo-annotators: the identification of claims is a hard task for human coders, so that even the merging of several independent human annotations does not guarantee full coverage. We found that relatively simple neural sequence classifiers were already good enough to substantially boost the recall during the merging phase (Padó et al, 2019).…”

“…In this section, we show how MARDY has been employed in a political science study targeting one of the major topics of German politics of 2015: the domestic debate on (im)migration policy. So far, we annotated 423 newspaper articles from TAZ (http://www.taz.de/), with a total of 982 claims (Padó et al, 2019).…”

“…This paper describes the MARDY tool, an interactive annotation environment for political claims analysis in computational political science (see Padó et al (2019) for a task analysis and initial modeling results). The term claim is operationalized as a textual span containing a demand, proposal, criticism, or a decision made by actors active in the respective field (Koopmans and Statham, 1999).…”

Do “Undocumented Workers” == “Illegal Aliens”? Differentiating Denotation and Connotation in Vector Spaces

“…Within NLP, such efforts tend to focus more on describing how language differs between political subgroups, rather than recognizing similarities in denotation across ideological stances, which is the primary goal of our work. For example, Preoţiuc-Pietro et al (2017); Han et al (2019) attempt to predict a person's political ideology from their social media posts, Sim et al (2013) detect ideological trends present in political speeches, Fulgoni et al (2016) predict political leaning of news articles, and Padó et al (2019) focuses on modeling the network structure of policy debates within society. Also highly related is work analyzing linguistic framing in news (Greene and Resnik, 2009;Choi et al, 2012;Baumer et al, 2015).…”

Swimming with the Tide? Positional Claim Detection across Political Text Types

“…The second line of research targets the direct automatic analysis of politic debates in textual form. As far as political claim analysis in newspaper articles is concerned, Padó et al (2019) have developed relatively simple embedding-based models for claim identification and classification. The experiments presented in this paper are based on their model architecture.…”