| 一种基于聚类的异常检测方法 |
| |
| 引用本文： | 李鑫,单维峰,丰继林,李军,高方平,李忠. 一种基于聚类的异常检测方法[J]. 传感器与微系统, 2011, 30(1): 19-21 |
| |
| 作者姓名： | 李鑫 单维峰 丰继林 李军 高方平 李忠 |
| |
| 作者单位： | 防灾科技学院,灾害信息工程系,河北,三河,065201 |
| |
| 基金项目： | 国家科技支撑计划资助项目，中国地震局教师科研基金资助项目，河北省教育厅自然科学研究计划资助项目 |
| |
| 摘 要： | 利用数据挖掘技术对网络中的海量数据进行分析从而发现入侵行为已成为目前异常检测研究的重点.为了进一步提高入侵行为检测的质量,提出了一种改进的异常检测算法.该方法首先将训练数据集转换为标准的单位特征度量空间,然后利用改进算法对数据进行划分,以找到聚类中心.最后对改进算法进行了性能分析与比较,实验结果表明:算法具有良好的稳定...
|
| 关 键 词： | 数据挖掘 聚类 异常检测 检测率 误警率 |
An anomaly detection algorithm based on clustering |
| |
| LI Xin,SHAN Wei-feng,FENG Ji-lin,LI Jun,GAO Fang-ping,LI Zhong. An anomaly detection algorithm based on clustering[J]. Transducer and Microsystem Technology, 2011, 30(1): 19-21 |
| |
| Authors: | LI Xin SHAN Wei-feng FENG Ji-lin LI Jun GAO Fang-ping LI Zhong |
| |
| Affiliation: | LI Xin,SHAN Wei-feng,FENG Ji-lin,LI Jun,GAO Fang-ping,LI Zhong(Department of Information Technology,Institute of Disaster Prevention Science and Technology,Sanhe 065201,China) |
| |
| Abstract: | Analyzing mass data in network by data mining technology in order to detect intrusion has become focus of anomaly detection research.In order to improve quality of intrusion detection,an improved anomaly detection algorithm is proposed.The training data set is converted to the standard unit features metric space.The improved algorithm is used to divide the data in order to find the clustering center.The improved algorithm is analyzed and compared with old algorithm.Experimental results show that the improve... |
| |
| Keywords: | data mining clustering anomaly detection detection rate false alarm rate |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
