XGB multi-dimensional operation and maintenance data anomaly detection method based on multivariate feature matrix

Abstract

The invention discloses an XGB multi-dimensional operation and maintenance data anomaly detection method based on a multivariate feature matrix, and the method comprises the following steps: constructing an METRIC table for collecting data according to product service data; collecting operation and maintenance data of the product service according to the METRIC table, and constructing a training sample; constructing a data set according to the training sample; training an XGB model by using the data set; and performing anomaly detection on the operation and maintenance data to be detected by using the trained XGB model. The multi-dimensional operation and maintenance data is subjected to anomaly detection for the first time, the METRIC table structure is designed through the multi-dimensional operation and maintenance data, the directional target of data monitoring and acquisition is achieved, the integrity of services is fully considered, the structural logic is complete, and productsupervision and data acquisition are better facilitated. Moreover, the XGB model is adopted for classification detection, and the XGB model is trained by comprehensively extracting data features and constructing a multivariate feature matrix, so that the reliability of the XGB model is improved, and the detection precision is further improved.

Description

technical field [0001] The invention belongs to the field of abnormal detection of intelligent operation and maintenance, and in particular relates to an abnormal detection method of XGB multi-dimensional operation and maintenance data based on a multivariate feature matrix. Background technique [0002] In recent years, high-tech applications such as mobile Internet, cloud computing, and big data are constantly maturing and evolving, requiring enterprises to keep up with the tide of the times, complete digital transformation, and avoid being eliminated by the times. The digital transformation of enterprises will bring them great opportunities for development, and at the same time, they will also face greater challenges. The IT environment and troubleshooting will become unprecedentedly complicated. In order to provide more efficient and rapid emergency service capabilities, more and more enterprises have begun to deploy intelligent operation and maintenance. In layman's te...

AI Technical Summary

Problems solved by technology

Due to the great difference in the threshold setting of different operation and maintenance data, the value range of some operation and maintenance data is extremely large. If different thresholds are set to deal with the problem of abnormal detection of multi-dimensional operation and maintenance data, the detection time will increase sharply. Moreover, If the operation and maintenance data are interfering (for example, the success and failure of the product business service are opposed to each other) it will also cause more serious false alarms

Therefore, threshold-based anomaly detection methods are extremely unsuitable for anomaly detection of multi-dimensional operation and maintenance data

Images