The development and adoption of natural language processing (NLP) methods by the political science community dates back to over twenty years ago. In the last decade the usage of computational methods for text analysis has drastically expandcd in scope and has become the focus of many social science studies, allowing for a sustained growth of the text-as-data community (Grimmer and Stewart, 2013). Political scientists have in particular focused on exploiting available texts as a valuable (additional) data source for a number of analyses types and tasks, including inferring policy positions of actors from tcxtual evidence (Lavcr et al., 2003; Slapin and Proksch, 2008; Lowe et al., 2011, inter alia), detecting topics (King and Lowe, 2003; Hopkins and King, 2010; Grimmer, 2010; Roberts et al, 2014), and analyzing stylistic aspects of texts, e.g., assessing the role of language ambiguity in framing the political agenda (Page, 1976; Campbell, 1983) or measuring the level of vagueness and concreteness in political statements (Baerg et al., 2018; Eichorst and Lin, 2018).
展开▼