Predict Attacker Behaviour on IDS with High Accuracy Using K-Nearest Neighbor Algorithm

This work’s main objective is to predict the attacker’s behavior pattern with high accuracy by using machine learning methods. According experimental and statical analysis, proposed model has improved accuracy. The study was performed two algorithms, K-Nearest Neighbor (KNN) Decision Tree (DTA). On a dataset of 19,864 items, algorithms were implemented, trained, assessed. Two iterations have extracted trained tested on sample size. Each algorithm undergone ten different test sizes get result sets. study’s G-Power for roughly 80%. sets programming experiment been further analyzed statistical tools observed that KNN 99.76, while DTA 98.84, according testing data. By conducting independent samples t-tests, difference p<0.05. research aims create an innovative intruder prediction uses techniques identify data as usual or invasive. While comparing decision tree algorithm, final results demonstrate outperformed.