Files
Abstract
Earlier attempts at understanding international alignments and alliance structures are often rigid in the sense that they prioritize a specific policy domain or time frame in international politics to present these alignments in easy to comprehend ways. Borrowing from the methodologies of Natural Language Processing, unsupervised machine learning, and social network analysis, I choose the United Nations General Assembly resolutions and voting data as a proxy for diplomatic alignment and show that different issue areas in international politics will yield alliance structures that are often in disagreement with the most widely used frameworks but align with our understanding of the dynamism of modern international politics. Computational approaches can provide a more precise and time-sensitive framework to think about countries’ alliance strategies.