A Novel Genetic Algorithm-XGBoost Based Intrusion Detection Method

In order to improve the speed and accuracy of model intrusion detection in complex network environment, a network intrusion detection method based on genetic algorithm-optimized XGBoost is proposed. Taking the NSL-KDD data set as the object, the XGBoost model is trained with the ten-fold cross valid...

Full description

Saved in:
Bibliographic Details
Main Authors: Sun, Yingying, Song, Chunhe, Yu, Shimao, Pan, Hao, Li, Tong, Liu, Yang
Format: Conference Proceeding
Language:English
Subjects:
classification
Complex networks
Computational modeling
Conferences
genetic algorithm
Machine learning
Network intrusion detection
NSL-KDD
prediction
Predictive models
Search problems
XGBoost
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In order to improve the speed and accuracy of model intrusion detection in complex network environment, a network intrusion detection method based on genetic algorithm-optimized XGBoost is proposed. Taking the NSL-KDD data set as the object, the XGBoost model is trained with the ten-fold cross validation method, and the genetic algorithm is used to optimize the model parameters to predict and classify whether the network is attacked. It not only avoids the problem of low classification accuracy of basic machine learning models, but also solves the problem of time consuming and low efficiency in conventional grid search. The experimental results show that compared with other machine learning classification models, the proposed model can not only improve the accuracy of detection, but also save the time cost and achieve a more ideal classification effect.
ISSN:2693-2776
DOI:10.1109/IMCEC51613.2021.9482357