Research on Network Malicious Code Immune Based on Imbalanced Support Vector Machines

作     者：LI Peng WANG Ruchuan 

作者机构：College of Computer Nanjing University of Posts and Telecommunications Jiangsu High Technology Research Key Laboratory for Wireless Sensor Networks 

出 版 物：《Chinese Journal of Electronics》 (电子学报(英文))

年 卷 期：2015年第24卷第1期

页      面：181-186页

核心收录：

学科分类：0808[工学-电气工程] 0809[工学-电子科学与技术（可授工学、理学学位）] 08[工学] 081104[工学-模式识别与智能系统] 081201[工学-计算机系统结构] 0811[工学-控制科学与工程] 0701[理学-数学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：supported by the National Natural Science Foundation of China(No.61170065,No.61373017,No.61203217) the Natural Science Foundation of Jiangsu Province(No.BK20140888,No.BK20130882) Scientific & Technological Support Project of Jiangsu Province(No.BE2012183,No.BE2012755) Natural Science Key Fund for Colleges and Universities in Jiangsu Province(No.12KJA520002) Scientific Research & Industry Promotion Project for Higher Education Institutions(No.JHB2012-7) Jiangsu Planned Projects for Postdoctoral Research Funds(No.1302090B) 

主　　题：Network malicious code immune Imbalanced support vector machines Computer network security Artificial intelligence Immunity theory. 

摘      要：The malicious computer code immune system and the biological immune system are highly similar:both preserve the stability of the system in real time in a constantly changing environment. This similarity is exploited to design a malicious code immune system to solve the malware active defense problem. The malicious code immunization project is mainly composed of four major components: the immune information collection program,immune information filtering processing program, immunization information discrimination program, and immune response program. An imbalanced support vector machine method was applied to optimize output results of malicious code immunization, thereby removing uncertain malicious code immune outputs. This demonstrates in detail the feasibility of the imbalanced support vector machine method in optimizing the immunization program output data. We showed that an imbalanced support vector machines can optimize the outputs of the malicious code immune system by removing glitches from the outputs. As a result,the machine helps to determine the precise time of the emergence of the immune response.