Structure Properties of Collaboration Network with Tunable Clustering

Long Wang,Guofeng Li,Yinghong Ma,Lu Yang

DOI: https://doi.org/10.1016/j.ins.2019.08.002

IF: 8.1

2020-01-01

Information Sciences

Abstract:A collaboration network model with improved triad formation (ITF) step is proposed and denominated as one-mode collaboration network with tunable clustering. We systematically study its structural features and find this network model can much more accurately reflect cooperative relationship in real society. We acquire expression of nodes degree distribution P(k) that possesses a power-law form with exponent gamma whose range is (1,3]. Expression of clustering coefficient C(k) for nodes with degree k is also obtained and composes of a power-law section and a constant section. We verify clustering coefficient of individual nodes increases monotonically with the increase of parameter p that is the probability of ITF step occurring. When specific forms of P(k) and C(k) are given, the model's average clustering coefficient C can be derived approximately. Finally, we analytically testify average path length L(N) increases with network size N in accordance with logarithmic form. When N is fixed, average path length of entire network decreases with the increase of p. All theoretically predicted results are consistent with simulative experimental results. (C) 2019 Elsevier Inc. All rights reserved.