Network intrusion detection method based on ambiguity and ensemble learning

Abstract

The invention discloses a network intrusion detection method based on ambiguity and ensemble learning. The method comprises the following steps: S1, from original traffic data, extracting basic characteristic data which can reflect traffic characteristics; S2, preprocessing the extracted basic characteristic data; S3, taking the preprocessed basic characteristic data as a training sample set, andrandomly dividing the training sample set for many times to obtain several training sample subsets; S4, according to data in each training sample subset, constructing and training several base classifiers through an ambiguity-based semi-supervised ELM algorithm; S5, combining all the well trained base classifiers through an integration algorithm, obtaining a final intrusion detection model, and performing network intrusion detection. With the method provided by the invention, problems, such as overlong training time for constructing the model and failure in well adaptation to detection of attacks of new types, in the prior art are solved, and the method provided by the invention can perform rapid training and improve capability of identifying the attacks of new types.

Description

technical field [0001] The invention belongs to the technical field of network intrusion detection, and in particular relates to a network intrusion detection method based on fuzzy degree and integrated learning. Background technique [0002] As a representative product of contemporary technology, the Internet has ushered in a new era of human society. The popularity and application of the Internet has also promoted the development of all aspects of human life, such as finance, education, medical care and so on. With the increasing number of Internet users, more and more attention has been paid to network security issues. Especially in recent years, various network attacks have been frequent, making it particularly important to establish a safe and reliable network environment. [0003] In order to effectively detect and defend against network attacks, the concept of Intrusion Detection System (IDS) has been proposed. The intrusion detection system detects whether there ar...

AI Technical Summary

Problems solved by technology

[0008] Aiming at the above-mentioned deficiencies in the prior art, the network intrusion detection method based on fuzziness and integrated learning provided by the present invention solves the problems in the prior art that the training time for building a model is too long and cannot well adapt to new types of attack detection

Images