The investigation of network structure has important significance tounderstand the functions of various complex networks. The communities withhierarchical and overlapping structures and the special nodes like hubs andoutliers are all common structure features to the networks. Network structureinvestigation has attracted considerable research effort recently. However,existing studies have only partially explored the structure features. In thispaper, a label propagation based integrated network structure investigationalgorithm (LINSIA) is proposed. The main novelty here is that LINSIA canuncover hierarchical and overlapping communities, as well as hubs and outliers.Moreover, LINSIA can provide insight into the label propagation mechanism andpropose a parameter-free solution that requires no prior knowledge. Inaddition, LINSIA can give out a soft-partitioning result and depict the degreeof overlapping nodes belonging to each relevant community. The proposedalgorithm is validated on various synthetic and real-world networks.Experimental results demonstrate that the algorithm outperforms severalstate-of-the-art methods.
展开▼
