基于正常轮廓更新的自适应异常检测方法

基金项目: 中南财经政法大学引进人才科研启动基金资助项目

详细信息

An Adaptive Anomaly Detection Method Based on Normal Profile Updating

¹ School of Information,Zhongnan University of Economics and Law,114 Wuluo Road,Wuhan 430060,China

Funds: 中南财经政法大学引进人才科研启动基金资助项目

摘要: 根据网络系统发生正常改变的基本特征,提出了确定网络系统正常改变的“三条件”计算方法,其计算结果可作为更新正常轮廓的依据。对正常轮廓的更新问题进行了深入探讨,提出了自适应异常检测的实现机制。并以网络流量分析为例,验证了在异常检测中应用这一方法的可行性。
- 异常检测 /
- 正常轮廓 /
- 规则更新
Abstract: According to the characteristics of legal change in Network,a three-premise computing approach is brought out to identify the legal change of protected Network.The problem of normal profile updating is discussed and the principle that designing an adaptive anomaly detection system is described.Using experiments on network traffic analysis,the feasibility of updating normal profile for anomaly detection system is validated.
- anomaly detection /
- normal profile /
- rules updating