Computational Text Analysis for Russian Language Media

This lecture covers text-as-data: from acquiring and preprocessing text, to building document-feature matrices, to the unique challenges of Russian morphology. It then introduces three topic modeling approaches (LDA, STM, and KeyATM) and two sentiment analysis methods: dictionary-based and transformer-based emotion detection. The examples are drawn from a 125,000+ article corpus of Russian and Western state-affiliated media coverage of the war on Ukraine.

Lecturer: Raushan Zhandayeva (https://www.zhandayeva.com), PhD, Postdoctoral Associate at the Institute for Data, Democracy, and Politics, GWU