Concept drift detection method based on classification error rate and consistency prediction

Abstract

The invention provides a concept drift detection method based on a classification error rate and consistency prediction, and belongs to the technical field of computer machine learning and informationsecurity. According to the method, mutation type concept drift is detected by calculating a change of the classification error rate of a model, and then the progressive concept drift is detected by calculating a consistency degree of the samples with wrong classification and the samples with correct classification so that the mutation type concept drift and the progressive concept drift can be detected in time, and relatively low calculation overhead is kept. According to the method, detection of mutation type concept drift and progressive concept drift is achieved at low calculation cost, and a model degradation phenomenon is recognized in time. The method is mainly used for concept drift detection, can effectively act on early judgment of a degradation phenomenon of a machine learning classification model, and can be used as a performance monitoring method in various application fields such as automatic analysis and decision in a big data environment.

Description

technical field [0001] The invention proposes a concept drift detection method based on classification error rate and consistency prediction, which belongs to the technical field of computer machine learning and information security. Background technique [0002] With the advent of the big data era, the application of machine learning algorithms in many fields has developed rapidly. Among them, information security based on big data technology is one of the most active research hotspots in recent years. A large number of machine learning classification models such as code clone detection models, malicious code classification models, vulnerability prediction models, and defect prediction models have emerged. These models are in High accuracy was achieved during construction. [0003] In the field of information security, software has evolved over time and different categories have emerged. Taking malicious software as an example, it can be divided into major categories such...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to solve the technical problems of concept drift detection of machine learning classification models in the field of software analysis in information security, and creatively propose a concept drift detection method combining classification error rate and consistency prediction, which is useful for evaluating current information security The performance and sustainability of the rapidly increasing security detection or evaluation systems based on machine learning classification algorithms in the field in the face of ever-evolving malware have important practical significance

Images