A Complex Network Model for Analyzing Railway Accidents Based on the Maximal Information Coefficient

It is an important issue to identify important influencing factors in railway accident analysis.In this paper,employing the good measure of dependence for two-variable relationships,the maximal information coefficient（MIC）,which can capture a wide range of associations,a complex network model for ra...

Full description

Saved in:
Bibliographic Details
Published in:理论物理通讯：英文版 2016, Vol.65 (10), p.459-466
Main Author: 邵福波 李克平
Format: Article
Language:English
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:It is an important issue to identify important influencing factors in railway accident analysis.In this paper,employing the good measure of dependence for two-variable relationships,the maximal information coefficient（MIC）,which can capture a wide range of associations,a complex network model for railway accident analysis is designed in which nodes denote factors of railway accidents and edges are generated between two factors of which MIC values are larger than or equal to the dependent criterion.The variety of network structure is studied.As the increasing of the dependent criterion,the network becomes to an approximate scale-free network.Moreover,employing the proposed network,important influencing factors are identified.And we find that the annual track density-gross tonnage factor is an important factor which is a cut vertex when the dependent criterion is equal to 0.3.From the network,it is found that the railway development is unbalanced for different states which is consistent with the fact.
ISSN:0253-6102