Anomaly based Intrusion Detection Model using Supervised Machine Learning Techniques

An intrusion detection system (IDS) is a software system that keeps track of network traffic and looks for anomalies. Abnormal or unusual network changes could be signs of fraud at any phase from the start of an attempt to a complete intrusion. Since data sharing primarily depends on the internet, it must be safe. For internet security, data encryption and authentication are insufficient and firewalls are unable to identify fraudulent packets that are fragmented. Moreover, attackers frequently vary their strategy, equipment, methods and tactics which can have disastrous results and effects such as lost productivity, financial loss, data loss etc. So, it become essential to put in place an effective intrusion detection system which is a very challenging task. The various supervised Machine Learning (ML) algorithms are applied in this paper, like J48, Random Forest, Random Tree, Hoeffding Tree and Logistic Model to predict the accuracy of an IDS system. The analysis was performed on the basis of three categories of data split and the algorithm that gives the best accuracy is suggested for future predictions. The various performance measures like accuracy, execution time, precision, F-measure and ROC curve are also analyzed. Random Forest exhibits best accuracy of 99.84% at a split ratio of 80:20 ratio as compared to other ML algorithms in all aspects. The execution time to build and test the model is less incase of Random Tree. As accuracy is the prime concern for an intrusion detection system (IDS); Random Forest is suggested to be the best solution as it provides highest accuracy.