MCA-based community detection

Abstract

In this work, we propose a new approach for consensus community detection based on MCA. The advantage of this approach is synthetizing the information coming from different methods and secondarily to obtain for each node relevant evidence about their different classification on more communities. This result can be important because the position of the single node can be interpreted differently from the other nodes on the community. In this way, it is possible to identify also different roles of the communities inside the network. The approach is presented and is shown by considering simulated networks and, at the same, time by considering some real cases of networks. In particular, we consider the real network related to the Zachary Karate Club.