You are here
Matje van de Camp
My interests lie in the extraction, visualization, and analysis of the information that is implicitly encoded in unstructured textual data. In my PhD research, I focus on the extraction of social networks. Many types of text (biographies, news articles, letters, diaries, novels, etc.) contain references to real-world or fictional entities, such as people, organizations, or places. By connecting all references to an entity across different documents and visualizing not only the entities mentioned in its vicinity, but also the context in which they are mentioned, we can create a multi-layered view that allows for the investigation of complex questions. For instance: how do ideas spread through the network? Who are the most influential actors? How do changes in the network relate to real-world events?
In solving the different steps of converting a source of flat text into a reliable dynamic resource for, in this case, historical research I try to stick to robust methods that are transferable to other domains with minimal adaptation.