How to connect with each other between roads? An empirical study of urban road connection properties

摘要: Abstract This paper aims to discover how connect with each other between roads from the perspective of degree–degreecorrelation in network science. Named street approach and stroke-like analysis were first combined define road cognition residents. Then, we constructed a series new measures characterize urban networks based on lane properties. We applied this method Xiamen, China. Based standard statistical physics, examined more detail distributions these found that original degree still conforms power-law distribution under definition. clear parity value classification was hierarchical measure, both them have good performance fitting. For improved adjusted degree, their whole rounding estimates are agreement Gaussian distribution, effect long tail is significant. In considering connection properties networks, no solo assortative or disassortative characteristic exists. ANND (average nearest neighbour degree) LC (link coefficient), they present distinct segmentation property different existing conclusions.