Nineteenth International Conference on New Directions in the Humanities

Abstract

The explosive growth of the Web as well as the increasing availability of content and resources, such as articles, blogs, publications, reviews, books create a heterogeneous and continuously increasing information reality that offers unlimited possibilities. Especially in the field of discourse analysis, the need to develop automated techniques for locating, extracting, and analyzing patterns and meanings in big textual data is of paramount importance. How can we utilize modern computational methods of text mining to identify concepts and track their semantic formation and their establishment as “common sense” in public discourse? How are specific ideological stereotypes being reproduced by the media? The proposed research attempts an empirical analysis of large textual data with innovative computational tools that process and query texts within a highly debated field: neoliberalism and the financialization of public discourse. We created a novel neoliberal corpus comprised by blogs and articles derived from the websites of four known and theory acknowledged institutions of neoliberal orientation and a body of classical neoliberal theorists which allowed for a twofold approach: a theory driven detection of structural elements of ideology and language and a data driven approach of feature extraction and domain specific concepts to create effective models and extract information that may elude the scrutiny of qualitative analysis and its small textual data. Our implementation of state-of-the-art algorithms and natural language processing libraries showed very promising results in exploring new ways social science may exploit big data.

Presenters

Katerina Mandenaki
Research Associate - PhD Candidate, Communication and Mass Media, National and Kapodristrian University of Athens , Attiki, Greece

Catherine Sotirakou
PhD Candidate, Research Associate, Faculty of Communication and Media Studies, University of Athens, Attiki, Greece