Applying The Fuzzy C-means Clustering Algorithm To Campus Network Security Assessment Based On The Complex Networks

With the development of computer technology and the growing network size, more and more attention on the network security is increasingly attracted by people. Similarly, the complex network, as a research tool for complexity science and complex systems, has aroused great interest in the field of computer network security research. Many national and international researchers have carried out high effective work and made great results. However, so far, most of the studies are also limited to network topology, network robustness, and computer virus and so on. At present, the ideas about the complex network have not seen yet to be introduced to the field of network security assessment study. For this reason, this paper takes a campus network as a study object of complex network, and uses an EAHP(Extension and Analytic Hierarchy Process)method to construct a campus network security evaluation index system. On this basis, it forms a fuzzy C means clustering algorithm based on weighted complex network features.In this paper, first, the definitions and the basic concepts of Extension and Complex network are briefly introduced to provide a theoretical basis; then, combined with suggestions of experts in the project team, the safety factors of the campus network security is analyzed, the campus network security evaluation index system is established, the EAHP is used empowering the indicators of evaluation system, and the wronging of common EAHP is analyzed. Again, the main campus network users are assessed in the way of a questionnaire method, and the relevant survey data are quantified for getting the results of the traditional comprehensive risk evaluation method. Finally, a new concept of risk conductivity is given, which, as a weighted side of adjacent nodes, two hypothesises is pointed that every node in a sample group transfers risk to other nodes and the campuse network is composed of several sample groups, and combined with the two major characteristics of the weighted complex networks (weighted degree coefficient and weighted clustering coefficient) and the fuzzy C means clustering algorithm, a fuzzy C means clustering algorithm with the weighted complex network features is formed to give the description of the clustering results and analysis.Based on the complex network theory, a quantitative assessment model of the network security is studied, which enriches the content of the network security assessment and the theory of complex network and presents a new thinking way of assessing network security.