Analysis of co-occurrence networks with clique occurrence information

Bin Shen, Innovation Centre for System Science and Big Data, Ningbo Institute of Technology, Zhejiang University, Ningbo, 315100, P. R. China
Yixiao Li, Corresponding author, School of Information, Zhejiang University of Finance and Economics, Hangzhou, 310018, P. R. China

Most of co-occurrence networks only record co-occurrence relationships between two entities, and ignore the weights of co-occurrence cliques whose size is bigger than two. However, this ignored information may help us to gain insight into the co-occurrence phenomena of systems. In this paper, we analyze co-occurrence networks with clique occurrence information (CNCI) thoroughly. First, we describe the components of CNCIs and discuss the generation of clique occurrence information. And then, to illustrate the importance and usefulness of clique occurrence information, several metrics, i.e. single occurrence rate, average size of maximal co-occurrence cliques and four types of co-occurrence coefficients etc., are given. Moreover, some applications, such as combining co-occurrence frequency with structure-oriented centrality measures, are also discussed.

Sunday, June 30, 2013
Monday, September 30, 2013
Sunday, December 8, 2013