Network Comparison on Structural Mining

Two common ways of network comparison are global property statistics and subgraph enumeration. However, they have unique traits, some advantages and some disadvantages. Global property of networks is easy to compute, but can only provide limited information; the subgraph-based method can extract essential structure of the studied networks, but for large networks, the exhaustible computing make subgraph enumeration practically infeasible. Here, we present a method to compare networks on structural mining. Based on their randomized network, deep features of the studied network emerge from the significant statistics. Experiments on yeast protein-protein interaction (PPI) networks and Internet at autonomous systems (AS) networks reveal their distinguished characteristics. The presented work provides a novel perspective for network comparison, also opens up new avenues for future research.