Identification of Overlapping Communities by Locally Calculating Community-Changing Resolution Levels

Abstract—The identification of thematic structures in networks of bibliographically or lexically coupled papers is hindered by the fact that most publications address more than one theme, which in turn means that themes overlap in publications. An algorithm for the detection of overlapping natural communities in networks was proposed by Lancichinetti, Fortunato, and Kertesz (LFK) last year [1]. The LFK algorithm constructs natural communities of (in principle) all nodes of a graph by maximising the local fitness of communities. The authors define fitness as the ratio of the number of internal links to the number of all links of the nodes of a community but the denominator of the ratio is raised to the power of α. This parameter can be interpreted as the resolution at which natural communities are determined. The resulting communites can, and are due to the constructing approach likely to, overlap. The generation of communities ca