Intrusion Detection on NF-BoT-IoT Dataset Using Artificial Intelligence Techniques
G. Aarthi, Shivam Priya, Wahab Aisha Banu · Advances in computational intelligence and robotics book series · 2023
The rapid development of internet of things (IoT) applications has created enormous possibilities, increased our productivity, and made our daily life easier. However, because of resource limitations and processing, IoT networks are open to number of threats. The network instruction detection system (NIDS) aims to provide a variety of methods for identifying the increasingly common cyberattacks (such as distributed denial of service [DDoS], denial of service [DoS], theft, etc.) and to prevent hazardous activities. In order to determine which algorithm is more effective in detecting network threats, multiple public datasets and different artificial intelligence (AI) techniques are evaluated. Some of the learning algorithms like logistic regression, random forest, decision tree, naive bayes, auto-encoder, and artificial neural network were analysed and concluded on the NF-BoT-IoT dataset using various evaluation metrics. In order to train the model for future anomaly detection prediction and analysis, the feature extraction and pre-processing data were then supplied into NIDS as data.