Text and Context in Understanding and Interpreting a State's Preferences through Text-mining of the UN General Assembly Speeches

Abstract

This research paper explores India's speeches in United Nations General Debate (UNGD) sessions to investigate the nation's priority settings and challenges faced during the last five decades. The study adopted Latent Dirichlet Allocation (LDA), a computational linguistics technique to uncover hidden topics in a selected corpus. The data were accessed from the United Nations General Debate Corpus. The research underpins Computational Grounded Theory, which relies on the computational exploration of the text. The results reveal that world peace, nation-building, poverty eradication, economic development, women empowerment, dialogue to resolve issues, and terrorism as the global threat are the main issues and challenges faced by India. Through topic modeling and word clustering, it becomes clear that India wants to eradicate poverty and terrorism through the joint effort of the whole world.