Cyberattack Detection Framework Using Machine Learning and User Behavior Analytics

This paper proposes a novel framework to detect cyber-attacks using Machine Learning coupled with User Behavior Analytics. The models the user behavior as sequences of events representing activities at such network. represented are then fitted into recurrent neural network model extract features that draw distinctive for individual users. Thus, can recognize frequencies regular profile manner in subsequent procedure is would abnormal by classifying unknown either or irregular behavior. importance proposed due increase especially when attack triggered from sources inside Typically detecting attacks much more challenging security protocols barely trustful resources network, including Therefore, be extracted and ultimately learned insightful patterns which reflect normal workflow. In contrast, trigger an alert potential cyber-attack. has been fully described where evaluation metrics have also introduced. experimental results show approach performed better compared other approaches AUC 0.97 was achieved RNN-LSTM 1. concluded providing directions future improvements.