Data anomaly intelligent detection based on machine learning

In order to distinguish whether it is abnormality with risk or without risk when environmental data collected by the device sensor is abnormal, this paper adpots k-means algorithm and hierarchical clustering algorithm in machine learning respectively to divide it, and compares the two algorithms on this research. The clustering results show that the risk-free abnormal data has the character of regularity, independence and contingency. The abnormal data with risk has the character of continuity and linkage.