基于加权支持向量机的网络入侵检测研究

在网络入侵检测中,数据类别不均衡训练集的使用将产生分类偏差,主要原因在于对每个训练样本的错误分类的惩罚系数是相等的.加权支持向量机对每个错误分类样本的惩罚系数是不一样的,这对小样本来说提高了分类精度,克服了常规SVM算法不能灵活处理样本的缺陷.但这是以大样本分类精度的降低以及总分类精度的下降为代价的.实验结果证明,将加权支持向量机用于网络入侵检测中是可行的、高效的.

Network intrusion detection based on weighted support vector machine

In the network intrusion detection,when the use of training sets with uneven class sizes results in classification biases towards the class with the large training size.The main causes lie in that the penalty of misclassification for each training sample is considered equally.Weighted support vector machines for classification where penalty of misclassification for each training sample is different,and then the classification accuracy for the class with small training size is improved,and overcomes the drawback which standard support vector machinen algorithm can not deal with this sample flexibly.But this improvement is obtained at the cost of the possible decrease of classification accuracy for the class with large training size and the possible decrease of the total classification accuracy.This introduce it to network intrusion detection,the experiment results prove it is effective and efficient.