摘要
针对目前网络入侵检测率低、误报率高的问题,提出一种基于半监督聚类云模型动态加权的入侵检测方法。由于属性对分类贡献程度不同,引入云相对贴近度的概念给出计算属性权重的方法。以半监督聚类算法为基础建立云模型,并对属性使用动态加权,通过对云模型的更新逐渐强化云分类器指导数据的分类。通过实验证明了该方法的可行性与有效性。
For the problems of low detection rate and high false alarm rate in current network intrusion detection, we propose a new intrusion detection method which is based on semi-supervised clustering cloud model and dynamic weighting. Since the attributes have different contribution to classification, by introducing the concept of ＂relative closeness degree of cloud＂ we give the attributes weight calculation method. We build the cloud model based on semi-supervised clustering algorithm, apply the dynamic weighting to attributes, and gradually strengthen the classification of the guidance data of cloud classifier by updating the cloud model. Through experiment it is proved that the method is feasible and effective.
出处
《计算机应用与软件》
CSCD
北大核心
2014年第3期322-324,共3页
Computer Applications and Software
基金
江苏省高校自然科学基金项目(05KJD52006)
江苏科技大学科研项目(2005DX006J)
关键词
半监督聚类
云模型
入侵检测
Semi-supervised clustering Cloud model Intrusion detection
作者简介
张杰，硕士生，主研领域：网络与信息安全。
李永忠，教授。
