Classification Model Of Io T Intrusion Detection Based On GWO-CatBoost

The Internet of Things(IoT)is a new model in computer networks.In the Internet of Things,various sensors are connected through various types of networks to achieve the ubiquitous connection of things to things and things to people.The main components of the Internet of Things are the Internet and various sensor networks.Due to the low computing power of sensors and their general exposure to natural environments,the Internet of Things is vulnerable to various network attacks.To overcome this problem,intrusion detection systems(IDS)play a key role as an active,high-security solution.This article proposes an IoT intrusion detection model based on GWO-CatBoost,and applies this model as a service to IoT security protection.The main content of this article is as follows:CatBoost is an integrated learning model with excellent performance and robustness.This article applies the CatBoost model to IoT intrusion detection.In order to verify the intrusion detection performance of the CatBoost model in the Internet of Things,we performed simulation experiments on the UNSW-NB15 dataset and selected support vector machine(SVM),random forest(RF),K-nearest neighbor(KNN),and Naive Bayes(NB)and other commonly used classification detection models for comparison.Simulation results show that our proposed model has better accuracy and stability of intrusion detection than the comparative model.Then we use the Gray Wolf Optimization(GWO)algorithm to optimize the parameters of the CatBoost model and propose a GWO-CatBoost-based IoT intrusion detection classification model.In order to verify the optimization ability of the gray wolf optimization algorithm and the intrusion detection performance of the GWOCatBoost model.We carried out simulation experiments on the UNSW-NB15 dataset with the three comparison models of GWO-CatBoost model and GWO-SVM model,GWO-KNN model,and GWO-RF.The experimental results verify the optimization ability of the GWO algorithm and show that the GWO-CatBoost model is significantly better than the comparison algorithm.