Simple Yet Powerful Machine Learning-Based IoT Intrusion System With Smart Preprocessing and Feature Generation Rivals Deep Learning

This project focuses on building an IoT intrusion detection system using machine learning to identify various attack types within IoT networks. It leverages the UNSW-NB15 dataset to classify attacks into eight categories: DoS, Exploits, Fuzzers, Generic, Normal, Reconnaissance, Shellcode, and Worms. The system applies several machine learning models, including Decision Tree (DT), Random Forest (RF), XGBoost, LightGBM, AdaBoost, Artificial Neural Networks (ANN), and Convolutional Neural Networks (CNN) for attack classification. These models are trained and evaluated to predict the nature of incoming attacks based on network traffic features. The system incorporates advanced preprocessing techniques to enhance the accuracy of the models by generating relevant features from the dataset. The backend is responsible for model training and prediction, while the frontend, developed using Flask, HTML, CSS, and JavaScript, provides a user-friendly interface. Users can register, log in, and input data to predict attack types, offering an effective solution for IoT network security management.

Keywords: IoT Intrusion Detection, Machine Learning, UNSW-NB15 Dataset, Attack Classification, DoS, Exploits, Fuzzers, Generic, Normal, Reconnaissance, Shellcode, Worms, Decision Tree, Random Forest, XGBoost, LightGBM, AdaBoost, Artificial Neural Networks, Convolutional Neural Networks, Feature Preprocessing, Flask, Web Application, Frontend Development, Backend Development, IoT Network Security.