Concept drift detection method and device

Abstract

The invention provides a concept drift detection method and device which are applied to flow detection. The method comprises the steps that electronic equipment obtains to-be-detected flow data and reference flow data for comparison, the to-be-detected traffic data is converted into a to-be-detected word segmentation vector, and the reference traffic data is converted into a reference word segmentation vector, so that whether the current traffic has concept drift or not is detected according to the similarity between the to-be-detected word segmentation vector and the reference word segmentation vector; therefore, an abnormal flow detection model can be updated when there is concept drift, and the problems that the accuracy is reduced and the false alarm rate is increased in the detection process of an abnormal detection method based on machine learning can be solved.

Description

technical field [0001] The present application relates to the field of flow detection, in particular, to a method and device for detecting concept drift. Background technique [0002] With the development of machine learning, the application of machine learning in the field of abnormal network traffic detection is gradually increasing, and it can accurately detect new unknown attacks in traffic. However, due to the concept drift of streaming data, that is, the style, meaning, and content of the data will change over time, so that when using machine learning methods for real-time network anomaly detection, the accuracy rate will change over time. Declining and increasing false alarm rates. Contents of the invention [0003] The purpose of the embodiments of the present application is to provide a concept drift detection method and device to solve the problems of decreased accuracy and increased false alarm rate during the detection process of anomaly detection methods base...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the embodiment of this application is to provide a concept drift detection method and device to solve the problem of accuracy in the detection process of the abnormality detection method based on machine learning The Problem With Declining and Rising False Positive Rates

Images