Journal of Beijing University of Posts and Telecommunications

  • EI核心期刊

JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM ›› 2015, Vol. 38 ›› Issue (2): 50-54.doi: 10.13190/j.jbupt.2015.02.008

• Papers • Previous Articles     Next Articles

Network Security Events Analyze Method Based on Neural Networks and Genetic Algorithm

LIU Jing1, GU Li-ze1, Niu Xin-xin1, YANG Yi-xian1, LI Zhong-xian2   

  1. 1. Information Security Center, Beijing University of Posts and Telecommunications, Beijing 100876, China;
    2. National Cybernet Security Ltd, Beijing 100088, China
  • Received:2014-04-01 Online:2015-04-28 Published:2015-05-14

Abstract:

The traditional network security events analysis methods depend more on human interventions. To address this problem, an automatic and self-adaptive method is presented. The neural network models are used to classify amounts of security events according to various attack scenarios, which can reduce much human intervention. The rule items are extracted from the classification results. And the correlation rules are generated automatically from these items using genetic algorithm. Experiments demonstrate that the method can classify the network security events and generate association rules automatically, so that the degree of automation can be improved. It is an effective enhancement and improvement to the traditional methods.

Key words: network security events analyze, neural network, generic algorithm, correlation rules

CLC Number: