Abstract

The problem of clustering large complex networks in the context of knowledge discovery and data management is central in current literature. In this paper, we present a method to unveil clusters present in complex networks, baptized COmplex Network CLUster DEtection (or, shortly, CONCLUDE). Our strategy relies on three steps: i) ranking edges centrality by using a random walker; ii) calculating the distance between each pair of connected nodes according to the ranking outcome; iii) partitioning the network into clusters so to optimize a function called network modularity exploiting both global and local information. The algorithm is computationally efficient since its cost is near linear with respect to the number of edges in the network. The adoption of our clustering method has been proved worthy in different contexts, such as for studying real-world social networks and artificially-generated networks with well-defined clusters.

Date

October 19, 2025

Authors

Emilio Ferrara, Alessandro Provetti