Enhancing IoT Network Attack Detection with Ensemble Machine Learning and Efficient Feature Extraction

Abstract

Attacks on Internet of Things (IoT) networks have been on the increase, making their security a top priority. Research in this area is on how to strengthen IoT network security via better attack detection using machine learning (ML) methods. To improve detection performance, this study proposes a novel method that combines effective feature extraction with ensemble ML methods such as XGBoost, LightGBM, and CatBoost. These algorithms are chosen due to their exceptional accuracy in classification tasks and their ability to handle complex, large datasets. The goal of the feature extraction procedure is to improve the efficacy and efficiency of the learning models by capturing essential properties from data collected by IoT networks. This will specifically design the model for IoT attack detection and train and validate it using a publicly accessible dataset from Kaggle. This study will use important metrics such as accuracy, area under the curve (AUC), precision, recall as well as F1-score to assess the effectiveness of the proposed approach in detecting assaults. The proposed method in this article is termed as IoT-SecureNet, which aims to enhance the security of IoT networks through the application of sophisticated ML techniques and the optimization of feature selection.