In the Digital Humanities, Network Analysis has been used to study first and foremost historical relations. Edges in these networks often represent social relations, or more abstract relations such as citation or co-appearance in a document. In most cases the nodes are human actors. More recently, Network Analysis software has been found helpful to also model texts and relations between them, as in the Stylo package for computational stylistics, which enables exporting results as edges for a network analytic study.
In the proposed presentation I wish to share the process and results of a short term scientific mission (STSM) undertaken in the framework of COST action “Reassembling the Republic of Letters”, during which I used the Tracer tool, a text reuse detection package developed by Marco Büchler, on the epistolary corpus of the project "Circulation of Knowledge and Learned Practices in the 17th-century Dutch Republic". Searching for a way to handle an overwhelming abundance of results, I turned to Network representation of the pairs of text reuse candidates.
This method not only enables assembling and making sense of the results of text reuse detection algorithms and recognizing different types of text reuse - quotes, aphorisms, formulae and factual statements; with attention to module characteristics given by the analysis a way opens to investigate how those types of text reuse behave in a communicative sphere. In fact, this method could be used in multiple research scenarios where textual phenomena are represented
As I will show, the remaining visualization challenge is to then superimpose the insights gained from the network analysis study of a textual phenomenon on a network - or other model - of the studied corpus.