The paper analyzed urban public transportation system by introducing some basic and important topological parameters such as degree distribution,average path length,clustering coefficient and betweeness and so on which were used in complex network analysis.

Some typical characteristics of the generated networks are carefully investigated through computer simulations,including the network diameter,the average path length,the degree distribution,clustering coefficient,degree correlation etc.

We characterize this complex network topology by measuring the clustering coefficient.

To address the issue in fuzzy C-means algorithm (FCM) that clustering number has to be pre-defined,a clustering algorithm,F-CMSVM (fuzzy C-means and support vector machine algorithm),is proposed for automatic clustering number determination.

Ascertainable clustering number and large training sets are vital problems of spectral clustering.

It presents a simple, easily implemented method on spectral clustering to automatically find the suitable value of clustering number and gives superior performance on experim.

In contrast to the old model,it has a much bigger cluster coefficient which is in conformity with the real worl.

Using analytical computing,we get the expression of degree distribution,cluster coefficient and average path length,and find this network model possessing complex networks characteristics like scale-free and small-.

Through analyzing in depth the present grey cluster model,we think they can be used to exact cluster with respect to cluster objectives when there are not distinguished differences between grey cluster coefficients,which often come forth in practical economic phenomena.