Computational Text Analysis and Big Data In the Post-Truth Era: Decoding Disinformation and Populism

he unprecedented volume of disinformation, misinformation, and populist rhetoric circulating online requires a methodological paradigm shift in the social sciences. To accurately decode these post-truth dynamics, researchers must increasingly rely on large-scale digital data analysis. Hosted by RC33 Logic and Methodology in Sociology, this session focuses on the application of computational text analysis and Big Data methodologies to uncover the narrative and cognitive dimensions of contemporary political manipulation.

While other analytical approaches might map network topologies, this session prioritizes the massive extraction and semantic analysis of digital content. We aim to understand how distorted messages are constructed, how populist emotional grammar operates at scale, and how these phenomena alter public discussion processes on controversial issues. We invite applied empirical works that employ Big Data techniques to trace the linguistic lifecycle of fake news across different digital platforms.

Specifically, we welcome abstract proposals utilizing Natural Language Processing (NLP), structural topic modeling, automated sentiment and toxicity analysis, and machine learning classifiers designed to detect deceptive content. By integrating these advanced computational methods, the session aims to foster interdisciplinary collaboration and significantly improve our analytical and explanatory capacity. Our ultimate goal is to equip sociologists with the robust, data-driven tools necessary to evaluate the discursive impact of post-truth dynamics on modern democracies.