摘要
根据分类技术建立入侵检测模型的思路,构造了一个基于贝叶斯分类的入侵检测原型系统。为了解决该方法存在的训练数据集问题,本文改进了现有的贝叶斯分类算法,提出了利用未标记数据提高贝叶斯分类器性能的方法。实验表明,这种方法取得了很好的效果。
According to the idea of designing intrusion detection system using classification, an intrusion detection system prototype is developed. In order to solve the problem existing in training data sets, present Bayes algorithm is improved and an algorithm using unlabeled data to improve the capability of the classifier is proposed. The experiment shows the result of this algorithm is good.
出处
《计算机科学》
CSCD
北大核心
2005年第8期60-63,共4页
Computer Science
