A data anomaly detection method, system and server having the system

Abstract

The present invention provides a data anomaly detection method, which is applied to electronic equipment. The data anomaly detection method includes the following steps: Step 1, based on the pre-stored single-stage support vector machine and the established operation data model, the decision-making corresponds to the operation data model The operating data of the hardware structure to distinguish the abnormal data and normal data in the operating data of the hardware structure; Step 2, vote on the identified abnormal data to determine whether there is an error marking the normal data as Abnormal data, if so, re-mark the operation data that is mistakenly marked as abnormal data as normal data. The present invention can achieve an accuracy similar to that of a system based on manual supervision, which is much higher than the accuracy of classic anomaly detection. Reduced server usefulness.

Description

technical field [0001] The invention belongs to the technical field of machine learning, and relates to a detection method and system, in particular to a data anomaly detection method, system and a server with the system. Background technique [0002] Anomaly detection is an important research direction in the field of machine learning. In the fields of finance, network security, etc., by learning a large amount of historical data, anomaly detection algorithms can distinguish normal data from abnormal data, so as to provide early warning for abnormal problems. Anomaly detection is usually tackled with support vector machine approach. see figure 1 , shown as a schematic plan view of the detection results of existing anomaly detection algorithms. Such as figure 1 As shown, the anomaly detection algorithm maps the detection data to a multi-dimensional space, and uses a hyperplane to separate normal data from abnormal data. Most of the training data of support vector machin...

AI Technical Summary

Problems solved by technology

[0005] In view of the shortcomings of the prior art described above, the purpose of the present invention is to provide a data anomaly detection method, system and server with the system, which are used to solve the problem that the existing anomaly detection technology cannot make real-time changes. When the actual hyperplane changes, concept drift will occur, which will lead to the problem of detection error

Images