Light Weight Gradient Ensemble Model for detecting network attack at the edge of the IoT network

An intelligent malware prediction system is needed for monitoring the network traffic of traditional and future network attacks. The Internet of Things (IoT) is fully connected and remotely monitored in many places. These devices are more vulnerable and easier to attack remotely. Securing and monitoring the activities of these devices must be done at the edge level. Machine Learning (ML) models are broadly used for supervising defence purposes, and these models are computationally expensive. Light Weight Gradient Ensemble Machine (LWGEM) with voting-based feature selection techniques and tuning the hyper-parameter are used to enhance the model accuracy and reduce the computational expenses. Performance of the model was evaluated with benchmark dataset UNSW_NB15 which shows the prominent result over the existing model and state of art models. The proposed model performs better in predicting unknown abnormalities in the IoT network. © 2022, The Author(s), under exclusive licence to Bharati Vidyapeeth's Institute of Computer Applications and Management.