AI Based Feature Selection for Intrusion Detection Classifiers in Cloud of Things

The popularity of cloud computing can be attributed to its on-demand nature, scalability, and flexibility. However, because of its heightened vulnerability and propensity for so-phisticated, widespread attacks, safeguarding this distributed en-vironment presents difficulties. Conventional IDS are insufficient. The proposed IDS for cloud environments in this study makes use of ensemble feature selection and classification techniques. This approach robustly distinguishes between attacks and normal traf-fic by merging individual classifiers through voting. Performance measures and ROC-AUC analysis show that the new approach is significantly more accurate and has fewer false alarms than the previous one. For cloud intrusion detection, this method provides a statistically better option.