Journal of Beijing University of Posts and Telecommunications

  • EI核心期刊

JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM ›› 2014, Vol. 37 ›› Issue (1): 71-75.doi: 10.13190/j.jbupt.2014.01.016

• Reports • Previous Articles     Next Articles

A K-Means Cluster Evaluation of Attack Effect Based on Bi-Dimensional Entropy Components

DAI Fang-fang1, ZHENG Kang-feng1, HU Ying2, LI Zhong-xian3   

  1. 1. Information Security Center, Beijing University of Posts and Telecommunications, Beijing 100876, China;
    2. Armed Police Beijing Command Academy, Beijing 100012, China;
    3. National Cybernet Security Ltd, Beijing 100088, China
  • Received:2013-04-12 Online:2014-02-28 Published:2014-01-07

Abstract:

A K-means cluster evaluation technique using bi-dimensional entropy components was proposed. The attack dataset on the basis of network entropy was preprocessed, a two-dimensional plane was mapped. The output of preprocess as the input of clustering was utilized. And a relation between the attack dataset and the effect category on the basis of K-means algorithm was established, thus an explicit division of attack effect set was achieved. Efficient evaluation was given. Experiment shows that the method can process attack dataset with high efficiency, as well as provide a visualized evaluation result by form of evaluation cluster diagram.

Key words: effect evaluation, clustering, entropy, K-means algorithm

CLC Number: