Screening node attributes that significantly influence node centrality in the network

5^th International Conference on Biometrics & Biostatistics

October 20-21, 2016 Houston, USA

WenJun Zhang

Sun Yat-sen University, China

Posters & Accepted Abstracts: J Biom Biostat

Abstract :

In the present study, I proposed a method to screen node attributes that significantly influencing node centrality in the network. One of four node centralities are degree centrality, closeness centrality, betweenness centrality and circuit centrality, can be used as the dependent variable and attribute-by-node data are used as the data of independent variables. Stepwise linear regression method was applied to screen statistically significant node attributes from candidate attributes. Matlab codes of the algorithm are provided.

Biography :

Email: wjzhang@iaees.org