Intrusion detection system based DTXG-RF machine learning performance against cyber attacks

Internet of Things (IoT) refers to physical objects connected by sensors, software and technologies and communicate data with other devices and systems over the Internet, IoT devices must be secure when hosted but when hackers discover new security weaknesses or flaws If they do, they can be sensitive. To maintain a strong security record and mitigate the risks of cyberattacks on their businesses, many companies have incorporated artificial intelligence (AI) into their security IoT systems. An intrusion refers to an unauthorized event or actions that affect the IoT ecosystem. Intrusion detection systems (IDSs) are critical to protect IoT systems from unauthorized access and maintaining their security integrity. This paper aimed to build an IDS utilizing a proposed ensemble machine learning (DTXG-RF) model which it is ensemble of three machine learning (Xgboost, Decision Tree and Random Forest) and comparing it with different machine learning models (DT, xgboost, LR, kNN, RF, NB, CATBOOST) for two datasets (IoTID20 and NSL-KDD) in terms of evaluation metrics like Precision, Accuracy, Recall, Fl-Score and confusion matrix. The results showed that the proposed DTXG-RF had outperformed other machine learning models with accuracy and reliability reached to 99.99% for the same datasets which improve the performance of the system.