Cyberattack Detection in IoT Networks Using Deep Learning

Abstract

This thesis presents a deep learning based approach for detecting cyber attacks in the Internet of Things(IoT)systems, mainly in the form of devices. Using the CICIoT2023 dataset, the study implements and compares Convolutional Neural Networks(CNN) and Long Short Term Memory(LSTM) models to identify many kinds of intrusions. This thesis also highlights the effectiveness of deep learning in handling large-scale IoT traffic data and improving the handling of large-scale IoT traffic data and improving detection accuracy, offering valuable insights for securing smart environments, including a brief analysis of traditional Machine Learning techniques like Decision Trees and Logistic Regression.