The project I explored was Using Metadata to Find Paul Revere by Kieran Healy. In this project, it is really exciting for me to see how social network analysis (SNA) can reveal hidden structures in data. By analyzing membership in colonial Boston organizations, this project demonstrates the power of metadata—information about connections rather than content—to identify key individuals and their roles in a network.
The nodes in this network are people (like Paul Revere and Samuel Adams) and organizations (such as the North Caucus and St. Andrew’s Lodge). The edges represent membership: a person is connected to an organization if they belong to it. This creates a bipartite network, where connections exist only between people and groups, not directly between individuals or organizations. However, the project goes further by using matrix multiplication to transform this bipartite network into two new networks: one connecting people through shared memberships, and another linking organizations through shared members. This duality highlights how individuals and groups are intertwined, revealing the social fabric of revolutionary Boston in a way that aligns with historical accounts.
The project was created using data manipulation and matrix operations. Starting with an adjacency matrix of people and organizations, the author multiplied the matrix by its transpose to create person-to-person and organization-to-organization networks. These transformations allowed the project to uncover patterns that weren’t immediately visible in the raw data. Centrality measures, like “betweenness centrality”, were then used to identify key figures. Paul Revere emerged as a central node, bridging multiple groups—a finding that aligns with his historical role as a connector in the revolution. His position in the network made him a critical figure, facilitating communication and coordination between different organizations.
What struck me most was how simple metadata—just membership lists—could uncover such rich insights. This has profound implications for modern applications, such as identifying influential figures in social media, mapping organizational structures, or even uncovering criminal networks. The project shows that even without detailed information about individuals’ actions or communications, the structure of their connections can reveal a great deal about their roles and influence.This capacity constitutes a big reason why I believe no matter what academic field it is, this type of analysis can have a huge potential in future investigation.
