Quantifying Global Networks of Exchange through the Louvain Method

A team of researchers has published a novel analysis applying network science to decades of U.S. policy documents. The study, "Quantifying Global Networks of Exchange through the Louvain Method," processed 2,010 Congressional Research Service (CRS) reports written between 1996 and 2024. By extracting mentions of countries and their interlinked interests, the team constructed a weighted graph representing 172 unique countries as nodes connected by 4,137 bidirectional edges. This creates a massive, data-driven map of international relationships as seen through the lens of U.S. congressional research.

To make sense of this complex network, the researchers applied the Louvain method, a greedy algorithm used to detect non-overlapping communities or clusters of countries with shared interests. They also computed eigenvector centrality for each country, a metric that identifies the most influential nodes—not just those with many connections, but those connected to other highly connected nodes. The result is a quantifiable ranking of global influence and a clear visualization of geopolitical blocs, moving beyond qualitative analysis to provide empirical evidence of how countries are interconnected based on policy interests.