Here’s a link to the project I’m writing about: https://kieranhealy.org/blog/archives/2013/06/09/using-metadata-to-find-paul-revere/
This network analysis project detailed the authors work on tracing Paul Revere’s “terrorist” connections through solely metadata. Usually this would involve looking into Paul Revere’s letters and/or methods of communication. However, the author uses membership data to form conclusions about key figures of the revolution.
Below is a dataset with names as individual observations and column variables, which are organizations at the time. The values are either 1, for active membership in the respective organizations, or 0, for not a member. The technique to create the network involves creating an adjacency matrix, which is a table with 254 rows and 7 columns that shows the connections between people and groups. By transposing the matrix and multiplying it with its transpose, a 254×254 “Person by Person” matrix is created, where the cells show the number of organizations any particular pair of people both belonged to.
From the network created above, it is clear each node is a person being documented and the edges are artificial relationship between them given that they both attend have a membership in the same organization. There isn’t any distinction between relationships (edges in the network).
The use of metadata to find “relationships” between individuals seems very useful in todays world. Think social media. The very participation of two people in the same online group could lead to data which motivates recommended profiles. The work done on this project is intriguing and definitely could result in more discoveries with more data. Think about the amount of data being collected on us on the daily through the web. Social media could easily reveal information about key players in a social network, the degree of connection between individuals or groups, and the overall structure of social relationships. However, the potential for network analysis to uncover valuable insights is significant, and as technology advances, it may become increasingly powerful in understanding the dynamics of social media and online communities.
