Cyberattack detection using hunter prey optimization based ML classification model

Recently, crime is becoming an observable way to make people and society under trouble. An imbalance in a nation’s population is caused by rising crime rates. Understanding the patterns of crime is essential to analyzing and responding to these kinds of criminal activity. Cybersecurity systems can evaluate and learn from patterns using Machine Learning (ML) models to identify attacks. Conventional security mechanisms fail in identifying sophisticated and emerging cyberattacks. To tackle this issue, this work presents a ML based cyberattack detection model that combines Hunter Prey Optimizer (HPO) for effective feature selection with for accurate classification. HPO algorithm selects the most essential features efficiently and improves the performance of the classification model. The Support Vector Machine (SVM) classifier exploits these features for detecting and categorizing cyber threats with better performance. The experimental analysis performed on the RTIOT2022 and attained better accuracy (99.1%), precision (98.9%) and recall (99.2%). The outcomes show the potential of the suggested model as a robust solution for smart cybersecurity systems.