摘要
特征提取是入侵检测的关键,特征提取的本质就是变换。针对传统核主成分分析(KPCA)在入侵检测应用中对模式分类问题所提取出的特征组合的不足,提出了一种基于信息度量改进的KPCA算法。数据集使用广泛地使用的KDDCUP99安全审计数据集,用训练样本的各特征向量的类内聚集程度和类间离散程度所组成的信息度量来代替传统KPCA中的累积贡献率,选取有利于检测出异常的特征组合。实验结果表明,改进的KPCA方法在较低的维数下就具有较明显的分类效果。
Feature extraction is the key to intrusion detection, transformation is the essence of the feature extraction. In view of the traditional kernel principal component analysis(KPCA) lack of extracted feature combination in intrusion detection application for pattern classification problems, this paper proposes an improved KPCA based on information measure. Data set uses the normative KDDCUP99 security audit data set. The aggregation degree within class and dispersion degree between class comprise information measure of each feature vector in the training sample. It is used to replace the cumulative contribution rate of the traditional KPCA. The selected feature combination is advantageous to detect abnormalities. A large amount of experimental results show that the improved KPCA method at a lower dimension will have a more pronounced effect of classification.
出处
《齐齐哈尔大学学报（自然科学版）》
2015年第4期1-3 9,9,共4页
Journal of Qiqihar University(Natural Science Edition)
基金
黑龙江省自然科学基金项目(F201218)
关键词
KPCA
特征提取
分类
入侵检测
KPCA
feature extraction
classification
intrusion detection
