论文部分内容阅读
该文在对网络连接数据进行分析和研究的基础上,将新超球面SVM方法运用于入侵检测中.在KDDCUP1999数据集上的实验结果表明,新超球面SVM入侵检测方法较之其他常用方法,体现出良好的分类性能以及检测性能.经过信息增益特征选择后,分类精度得到进一步提高.“,”This paper presents a new intrusion detection method based on quarter-sphere support vector machine after analyzing and studying with the network connection data. Experimental results carried out on KDD CUP1999 data set after feature selection using Information Gain show that the proposed method performs better both in true positive rate and precision accuracy compared with other common intrusion detection methods.