Machine Learning Mechanisms for Network Anomaly Detection System: A Review

Network Anomaly Detection Systems (NADS) has a great importance in Network Defense System for detecting potential or critical threats. Numerous Organization have actualized, Intrusion Detection System (IDS) as a security segment, and introduced the various mechanism to recognize the effect of the system assaults. However, Machine Learning methods are widely used in IDS to detect the various attacks. In this context, network traffic dataset plays very important role. Hence, IDS uses those datasets to learn about normal and anomalous activities. Whereas the labelled datasets are used for training phase. As appropriate selection of Machine Learning methods gives the better result, therefore, a comparative study about few machine learning methods have been used in this article using NSL-KDD dataset for the analysis purpose. Finally, the simulated results have been compared by implementing of Naïve Bayes classifier (NB), Support Vector Machine (SVM) and Decision Tree classifier on NSL-KDD dataset. Recursive Feature Elimination (RFE) and Principal Component Analysis (PCA) have been used for selecting the appropriate features among all features present in the dataset to improve the accuracy and processing speed of the IDS.