Proceedings of the 8th International Academic Conference on Research in Social Sciences
Year: 2024
DOI:
[PDF]
Polarization in Online Public Discourse: Analyzing Twitter Conversations in Post-2001 Afghanistan
Abdul Razaq Vahidi
ABSTRACT:
Afghanistan has witnessed increasing polarization in the online public discourse in the post-2001 period, highlighting a widening division of perception over critical social and political issues. This research explores the evolving socio-political polarization trends and aims to analyze and categorize Twitter conversations using keyword frequency analysis, clustering algorithms, and trend analysis over time. We use topic modeling in Natural Language Processing (NLP) with Machine Learning (ML) to analyze positive, negative, and neutral sentiments across predefined themes such as governance, gender, social justice, terrorism/security, and culture. The study reveals significant shifts in the focus of Twitter conversation with varying emphasis on different themes and the interrelation of discussions over time intervals from 15 August 2021 to 9 May 2023. As the first study to pinpoint central issues in Afghanistan’s public discourse and polarization trends in the online public sphere, this study highlights key trends and thematic clusters. It offers a timely contribution to understanding the complex dynamics of political discourse in Afghanistan.
keywords: Education, Public Discourse, Security, Sentiment Analysis, Social Justice