摘要
GECISM(GEneral computer immune system model)是基于规则匹配检测的计算机免疫系统,免疫识别规则对“自我”和“非我”特征的表征能力直接影响到GECISM的性能,所以挖掘高效免疫识别规则的是GECISM的一个重要研究内容。改进后的Apriori算法以系统调用序列为数据源,从“自我”集和“非我”集中计算出频繁谓词,进而产生免疫识别规则。这些规则反映了“自我”和“非我”的内在特征,是GECISM进行“非我”检测的判据。
GECISM is a rule for matching detection based computer immune system whose performance directly relates to how much these recognition rules can represent the features of self and nonself,so extracting effective recognition rules is the key theme to implement GECISM system. Revised Apriori algorithm calculates frequent predicates from system call sequence which is the raw data, and then extracts recognition rules. These rules reflect some inherent feature of self and nonself,and they are the basic detection criterion of GECISM.
出处
《广西师范大学学报（自然科学版）》
CAS
北大核心
2007年第2期38-42,共5页
Journal of Guangxi Normal University:Natural Science Edition
基金
Plan of Research on Science and Technology and Development in Hebei Province(04213534,04213529)
