Intrusion Detection with Ant Colony Optimization Based Feature Selection and XGboost Classifier
B Shweta, Rawat Seema, Bindu Maringanti Hima · International Journal of Performability Engineering · 2024
The Internet of Things is being used far more frequently.Malicious attacks are on the rise in tandem with the growth of technology such as smart devices, smart homes, and other forms of automation.Taking care of network security is crucial.An effective network intrusion detection system is essential for defending Internet of Things devices against malware.It consists of feature selection, classification, and dimensionality reduction.Principal component analysis (PCA) is used to reduce dimensionality, and ant colony optimization (ACO) is used to choose features.Following that, the Extreme Gradient Boost Algorithm (XG-Boost) is used to carry out the classification.Python is used to implement the benchmarked intrusion detection NSL-KDD dataset on Colaboratory.The system outperformed the current intrusion detection system with 99.27 % accuracy, 99.6% precision, 98.8% recall, and 99.2% F1-score.Based on testing results, the system seems to be robust when evaluating several performance criteria.