A Novel Cutting Double k-Clique Spanning Tree Method for Coalition Detection from Brain Functional Networks

Network science is a useful tool to investigate brain functional networks. In particular, community detection could be considered as a representative framework to identify brain functional coalitions. However, brain functional networks generally demonstrate dense structure feature and imbalanced coalition scale. It causes high-complexity and imbalance problems for community detection. In this paper, a novel method, named Cutting Double k-Clique Spanning Tree (COUSIN) method, for coalition detection from brain functional networks has been proposed. The local nodes are characterized into k-cliques as essential substructures. Then the k-cliques with each degressively order k hierarchically merge and form two spanning trees of the given network. A cutting strategy is used to drop coalitions from the double spanning tree. Experiments of robustness verification are executed on simulated networks. The experimental tests on real datasets exhibit prominent performance of community detection of the proposed method. The obtained coalitions from actual brain functional networks indicate the application value of the method in neuroscience.