No abstract

Intrusion detection has become an indispensable technique to ensure the security and reliability of information systems. Support vector machine (SVM) and its many improved algorithms have been successfully applied to intrusion detection systems in recent years. However, the training process of SVM ignores an important prior knowledge, the within‐class structure in the training set. In this paper, we propose an improved classification algorithm, which combines minimum within‐class scatter in Fisher discriminant analysis with traditional SVM. The central idea is to find an optimal separating hyperplane such that the margin is maximized, whereas the within‐class scatter is kept as small as possible. This new algorithm is called SVM with minimum within‐class scatter (WCS‐SVM). A set of experiments is conducted on ten benchmarking datasets and KDDCUP'99 experimental data of MIT Lincoln Laboratory to test the generalization performance of the WCS‐SVM algorithm. Experimental results show that the WCS‐SVM algorithm has better discriminatory power than traditional SVM and kernel Fisher discriminant analysis and it has higher true detection rate and lower false positive rate for intrusion detection systems. Copyright © 2012 John Wiley & Sons, Ltd.

Outlier detection, as a type of one-class classification problem, is one of important research topics in data mining and machine learning. Its task is to identify sample points markedly deviating from the normal data. A reliable outlier detector needs to build a model which encloses the normal data tightly. In this paper, an improved one-class SVM (OC-SVM) classifier is proposed for outlier detection problems. We name this method OC-SVM with minimum within-class scatter (OC-WCSSVM), which exploits the inner-class structure of the training set via minimizing the within-class scatter of the training data. This can construct a more accurate hyperplane for outlier detection, such that the margin between the training data and the origin in a higher dimensional space is as large as possible, while at the same time the decision boundary around the normal data is as tight as possible. Experimental results on a synthetic dataset and 10 real-world datasets demonstrate that our proposed OC-WCSSVM algorithm is effective and superior to the compared algorithms.

No abstract