Similar to text mining and data mapping, network analysis is a method of visualization which allows researchers to demonstrate relationships between information, especially on a large scale. Oftentimes, humanists will upload hundreds or even thousands of individual data points into their network software which are referred to as “nodes”, and additional datasets which connect between the nodes become “edges.” Initially, the results of graphing data sets at this scale can be quite complicated to understand, however programs such as Gephi have built-in tools to enable users to detangle this web of information in a way that makes sense.
Once the network analysis has been unraveled, certain trends will begin to emerge through particular grouping or “clusters” of data, revealing patterns between groups or individuals who share characteristics with one another. Depending on the type of data a researcher is working with, these clusters will have varying levels of significance for their work. For instance, the Linked Jazz project network analysis uses gender as one of its categories for organization in order to demonstrate the importance of prominent female musicians on the jazz scene as a whole. Humanists have already accustomed themselves to think about their datasets as interwoven patterns of connections, however, it was only recently that they have developed the necessary software to convey this way of thinking to a more general audience.
http://maximiliansmith.com/gephi/bimodal
Leave a Reply