Anomaly Detection in Network Traffic Using Machine Learning Techniques

Abstract

It has been nothing less than exponential growth in the last two decades, and with this growing Internet has come unprecedented connectivity, significantly increasing the number of cyberattacks. Zeroday attacks have always challenged traditional signature-based detection techniques, which is why anomaly-based detection techniques have become increasingly important for identifying any anomalies in normal network behavior. Key features were selected using the Random Forest Regressor. Seven machine learning algorithms are tested in this experiment.