Can network analysis reveal importance? Degree centrality and leaders in the EU integration process

Abstract

This paper describes ongoing work on the potential of simple centrality algorithms for the robust and low-cost exploration of non-curated text corpora. More specifically, this paper studies (1) a network of historical personalities created from co-occurrences in historical photographs and (2) a network created from co-occurrences of names in Wikipedia pages with the goal to accurately identify outstanding personalities in the history of European integration even within flawed datasets. In both cases Degree centrality emerges as a viable method to detect leading personalities.