Abstract
In this paper, we have designed and implemented a solution for the detection of cyber threats using supervised machine learning (ML). An effective software program was adapted and refined in Python Language for training our machines on a large cyber-security dataset in order to detect and classify various types of network intrusions, and make use of features extracted from network traffic to identify known intrusion attacks. The intrusion detection system IDS with the usage of ML is evaluated using standard metrics such as accuracy, precision, recall, and Fl-score. Four different models were used in this study, namely Decision Tree, Random Forest, State Vector Classifier, and KNN Classifier. The performance of these models after training them on a large dataset was tested on the same data set and on a new data set. These tests showed very promising and satisfactory results.