Evaluation of Various Machine Learning Algorithms for Detection of Attacks in 5G

Security plays an essential role in IoT as there are several components connected to the network. An effective intrusion detection system is required to detect all vulnerabilities in wireless networks using machine learning and also to reduce the false alarm rate. The various machine learning algorithms were analyzed in terms of various performance metrics and finally evaluated the best classifier in terms of prediction time, training time, and accuracy. Using confusion matrix, various classes of attacks were analyzed, and results revealed that various machine learning algorithms were compared, And our proposed work was able to predict all the five types of attacks effectively with an accuracy of 94% achieved using extra tree classifier.