An efficient community detection method based on rank centrality

A graph theoretic approach for unsupervised feature selection

“…The classical clustering approach, k-means, has been shown to be very efficient to detect communities in networks. However, k-means is quite sensitive to the initial centroids or seeds, especially when it is used to detect communities (Shi et al, 2013;Jiang et al, 2013). The goal of community detection is to cluster the similar vertices into one community separate from others.…”

A parameter-free community detection method based on centrality and dispersion of nodes in complex networks

“…In case that community structure of a network is fuzzy, it is very difficult to identify initial seeds positioned distinctly in the decision graph, we estimate the number of communities K by F statistics [35,39,48] to find out the top K nodes with high comprehensive value as initial seeds. Comprehensive value (CV) of node V i is defined as follows,…”

“…K-rank [35] is a representative vertex clustering algorithm for community detection, which takes node connectivity of a network into account to select initial seeds. The initializing procedure is comprised of two steps.…”

An Extreme Learning Machine‐Based Community Detection Algorithm in Complex Networks

“…k-means++ chooses distinct initial seeds far from each other in a probabilistic manner, which leads to more stable clustering results but involves increased complexity [26]. Through ranking nodes in the same manner as Google's cofounders did [27] and picking the center nodes from the highest ranking ones, k-rank achieves small fluctuation in the community detection output although it requires additional running time [28]. Another defect of k-means, explained by Ng et al [29], is that it is only capable of finding clusters corresponding to convex regions.…”