Multiclass Email Classification by Using Ensemble Bagging and Ensemble Voting

Email is a common communication technology in modern life. The more emails we receive, the difficult and time consuming it to sort them out. One solution overcome this problem create system using machine learning emails. Each method of data sampling result different performance. Ensemble combining several models into one model get better In study tried multiclass email classification by models, sampling, classes obtain effect Bagging Voting methods on performance macro average f1 score, compare with non-ensemble models. results show that sensitivity Naïve Bayes imbalance helped ∆P (delta performance) range 0.0001 – 0.0018. Logistic Regression has 0.0001-0.00015. Decision Tree lowest compared others -0.01