Method and device for constructing pu learning-based malicious traffic detection model

Abstract

The invention discloses a method and device for constructing a malicious flow detection model based on PU learning, relates to the technical field of networks, and mainly aims to construct a maliciousflow detection model based on machine learning. According to the main technical scheme, the method comprises the steps of obtaining flow data as a sample data set; training based on the sample data set to obtain a plurality of candidate models; Constructing an evaluation set based on the sample data set; Evaluating each candidate model according to the evaluation set and a preset evaluation condition to obtain an evaluation result corresponding to each candidate model; Selecting a candidate model with an evaluation result meeting a preset condition; And integrating the selected model according to a preset integration method to obtain a malicious flow detection model. The method and the device are used for realizing the process of constructing the malicious flow detection model in the malicious flow detection process.

Description

Technical field [0001] The invention relates to the field of network technology, in particular to a method and device for constructing a malicious flow detection model based on PU learning, and a method and device for malicious flow detection. Background technique [0002] With the continuous development of network technology and people's work and life are more and more closely related to the network, the traffic on the network has gradually increased. Among them, malicious traffic often exists in the network, which will affect the normal operation of the website database or system, such as the network. Attacks, traffic fraud, malicious crawlers and other network traffic are common malicious traffic. Such malicious traffic usually invades, interferes with, or captures unauthorized business data or information through unauthorized methods. Based on the harmfulness of malicious traffic, experts in the domain have paid more and more attention to the detection of malicious traffic. ...

AI Technical Summary

Problems solved by technology

Therefore, based on the existing detection methods, it is difficult to meet the current needs of malicious traffic detection. How to realize a detection of malicious traffic based on machine learning has become an urgent problem to be solved in the industry.

Images