不同SVM算法在入侵检测中的应用

近年来支持向量机在入侵检测领域得到了广泛应用,由于支持向量机理论在发展过程中不断涌现出新的算法,为了找到一种较适合入侵检测的算法,选择了具有代表性的基于C-SVM的SMO算法和一种新的支持向量机LS-SVM,分别应用于入侵检测.使用不同规模训练集和测试集进行多组实验,从不同角度研究了它们在入侵检测中的特性,并进行综合比较研究,从实时性、检测精度、误报率和漏报率方面研究它们在入侵检测中的优劣,找出较优的算法为SMO算法.

The Application Different SVM Algorithm in Intrusion Detection

Support Vector Machine has been widely used in the field of Intrusion Detection in resent years, and new algorithms are emerging out of in its development process. In order to find a more suitable SVM algorithm for Intrusion Detection, the representative SMO algorithm based on C-SVM and a new support vector machine LS-SVM were chosen separately and applied to Intrusion Detection, with different size training set and test set for multiple experiments to find their properties in Intrusion Detection from different perspectives and comparative comprehensively. Study their advantages and disadvantages in Intrusion Detection from real-time, detection accuracy, false positive rate and false negative rate,and find the SMO algorithm, as the optimum algorithm.