Author ORCID Identifier
Semester
Fall
Date of Graduation
2025
Document Type
Dissertation
Degree Type
PhD
College
Eberly College of Arts and Sciences
Department
Political Science
Committee Chair
Erik Herron
Committee Member
Matthew Jacobsmeier
Committee Member
Mason Moseley
Committee Member
Vito D'Orazio
Committee Member
Megan Vendemia
Abstract
This dissertation explores how competing interests across secondary dimensions of identity shape marginalized groups' capacity to institutionalize their grievances, particularly ethnic minorities. More specifically, I explore two mechanisms, intragroup cohesiveness and distinctiveness, as a driving force of grievance mobilization. This research is interdisciplinary, drawing on theories from political sociology, political communications, and computational social science. I examine the question of ethnic group particization in two stages. In the first stage, I developed an original machine learning pipeline to process large text corpora and identify the indicators of the strength of collective ethnic grievances in online group discussions. In the second stage, I collected a cross-national dataset comprising over 100 ethnic minority groups across 21 post-communist countries to measure their socioeconomic and linguistic cohesiveness and distinctiveness. I find that the two mechanisms, intragroup cohesiveness and distinctiveness, are essential for explaining the mobilization capacity of ethnic minorities to institutionalize their grievances.
Recommended Citation
Tepnadze, Ani, "Application of Machine Learning and Natural Language Processing to Understand the Partisan Mobilization of Ethnic Minorities" (2025). Graduate Theses, Dissertations, and Problem Reports. 13074.
https://researchrepository.wvu.edu/etd/13074
Included in
Communication Technology and New Media Commons, Comparative Politics Commons, Gender, Race, Sexuality, and Ethnicity in Communication Commons, Race and Ethnicity Commons, Social Psychology and Interaction Commons
Comments
the Libraries Committee Signature form