From the study of the two major characteristics of complex networks as a starting point, this paper presents a risk-conductivity as a weight side between neighbor nodes to construct a weighted complex network. Further, the concepts of weighted degree coefficient and weighted aggregation coefficient as well as their computing methods are given, and for the cluster analysis, a weighted fuzzy C-means clustering algorithm based on the weighted complex network features is formed by putting the two characteristic values. Finally, this method applied to assess a campus network security is verified by doing an experiment.