A PSO and Random Forest-based Hybrid Model for Effective Intrusion Detection

Recent technological advances have led to the need to store information electronically, which is why information security has become an increasingly important issue in recent years. Traditional intrusion detection takes care of unusual activity. It is crucial to improve the true positive while reducing false positives to having a trustworthy IDS in a network. We present a methodology intrusion detection system (IDS) that relies on machine learning classifiers and multi-objective PSO in our proposed model. The KDDCUP99 dataset is picked up as an input for evaluating the outcome of the proposed model. Based on the performance analysis and the comparative study, it can be concluded that the model is valid. PSO-RF proposed model gives better classification accuracy compared with DT, KNN, SVM, and Naïve Bayes and the highest Gbest value with maximum TPR and minimum FPR.