Community detection using k-means clustering based on top influential nodes identified by TOPSIS method
This project aims to study a complex network by identifying its top-k influential nodes by a proposed method and detecting the communities associated with these nodes using k-means clustering. The first task focuses on using the proposed method TOPSIS in order to identify the top-k influential nodes. These nodes will be compared to top-k influential nodes identified by the different centrality measures, DC (degree centrality), BC (betweenness centrality), CC (closeness centrality) and EC (eigenvector centrality), using an evaluation model SI (Susceptible, Infected). The objective of the second task is to use the k-means clustering algorithm in order to determine k clusters based on the top-k influential nodes identified by TOPSIS and to compare its result with other community detection algorithm.