Currently, there is a growing interest in identifying communities in social networks. Although there are many algorithms that suitably resolve this problem, they do not properly find overlaps among communities. This paper describes a new approach to the detection of overlapping communities based on the ideas of friendship and leadership, using a new centrality measure, called extended degree. We describe the algorithm in detail and discuss its results in comparison to CFinder, a well-known algorithm for finding overlapping communities. These results show that our proposal behaves well in networks with a clear leadership relationship, in addition it not only returns the overlapping communities detected but specifies their leaders as well. © 2011 Springer-Verlag.
CITATION STYLE
Palazuelos, C., & Zorrilla, M. (2011). FRINGE: A new approach to the detection of overlapping communities in graphs. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6784 LNCS, pp. 638–653). https://doi.org/10.1007/978-3-642-21931-3_49
Mendeley helps you to discover research relevant for your work.