“…Over the past few years, a number of models and approaches based on traditional machine learning have been proposed for network intrusion detection. Examples include the Support Vector Machine (SVM) [24], [25], K-Nearest Neighbors (KNN) [26], Artificial Neural Networks (ANN) [17], [18], Random Forests (RF) [24], [25], [59], Decision Trees (DT) [23], [26], [27], [60], [61], Linear Regression (LR), Naïve Bayes (NB), Expectation Maximization (EM) [23], Simulated Annealing (SA) [27], Simplified Swarm Optimization (SSO) [28], Neutrosophic Logic (NL) [29], Neurotree [30], Random Effects Logistic Regression (RELR) [31], PCA filtering [62], and others reported in [32]- [34]. Recently, Nawir et al [35] proposed a classification model based on the Average One Dependence Estimator (AODE) algorithm for multiclass classification, on the UNSW-NB15 dataset, reporting an accuracy of 83.47% and a false alarm rate (FAR) of 6.57%.…”