Time series data anomaly detection method and device, equipment and storage medium

Abstract

The embodiment of the invention provides a time series data anomaly detection method and device, equipment and a storage medium, and the method comprises the steps: obtaining a to-be-detected time series data fragment and a plurality of historical time series data fragments, the to-be-detected time series data fragment being a time series data fragment with numerical value mutation; calculating the Euclidean distance between each historical time sequence data segment and the time sequence data segment to be detected; judging whether a historical time sequence data fragment exists, wherein the Euclidean distance between the historical time sequence data fragment and the to-be-detected time sequence data fragment meets a distance similarity condition; and if not, sending an abnormal alarm for the to-be-detected time sequence data fragment. If there is no historical time series data fragment close to the time series data fragment, it is indicated that numerical value mutation of the time series data fragment does not occur in a short time and cannot be determined to be caused by a fixed behavior of a user in a short time, and under the condition, it is judged that the time series data fragment is an abnormal time series data fragment, and then an abnormal alarm is sent; and invalid alarms for time series data exception can be reduced.

Description

technical field [0001] The present invention relates to the technical field of data analysis, in particular to a time series data anomaly detection method, device, equipment and storage medium. Background technique [0002] In the Internet platform, a large amount of time-series data is generated every day. Time-series data refers to time-series data, which is a data column recorded in chronological order by the same unified index. Each data in the same data column must be of the same caliber and comparable . [0003] Time-series data anomaly detection is an auxiliary means for judging data status. The purpose is to find abnormal data in time, facilitate timely measures to eliminate or weaken the anomaly, reduce the loss caused by the anomaly, and ensure the normal operation of various businesses. [0004] In the existing technology, the upper and lower thresholds of the time series data in the future can be predicted based on the historical time series data, and then anoma...

AI Technical Summary

Problems solved by technology

It can be understood that the fixed behavior of users in a short period of time is a relatively common behavior pattern. Therefore, there is no need to pay attention to the phenomenon that the value of time series data frequently increases rapidly in a short period of time. In this case, a large number of invalid Alarm, which increases the operation and maintenance cost of time series data

Images