Abnormal Detection Research Based on Outlier Mining

This paper introduces the latest development of intrusion detection based on data mining, and studies the key technologies of network intrusion detection based on data mining, especially outlier mining. A anomaly detection model based on outlier mining is proposed. The model firstly preprocesses the data and sets the normal sample set, calculates the outlier score of each sample in the normal sample set, and combines it with the specified outlier value threshold. The normal behavior profile is formed, which can be used as the basis for detection.