A Comprehensive Analysis of Intrusion Detection System using Machine Learning and Deep Learning Algorithms

The Internet of Things (IoT) plays an essential character in both domestic and corporate sectors by facilitating connectivity and automation through a network of devices and sensors. IoT wearable gadgets, sensors, and embedded devices are widely utilized for data collection and transmission. However, privacy and security are challenging tasks to be achieved while the standard implementation of monitoring and security operations is not possible because of large number of IoT devices, their heterogeneity and technical specification limitations. In order to solve this issue, Intrusion Detection System (IDS) is introduced to secure the network against various vulnerabilities and threats. To identify intrusion attacks accurately, Machine learning (ML) and Deep Learning (DL) algorithms namely, Random Forest (RF), Logistic Regression (LR), Decision Tree (DL), AdaBoost, XGboost, Convolutional Neural Network (CNN), Deep Convolutional Neural Network (DCNN), and Long Short-Term Memory (LSTM), Deep Neural Network (DNN) are used. This research aims to identify the strengths, relationship, and challenges of the ML and DL-based prediction approaches.