Model optimisation of class imbalanced learning using ensemble classifier on over-sampling data

<span lang="EN-US">Data imbalance is one of the problems in application machine learning and data mining. Often this occurs most essential needed case entities. Two approaches to overcome problem are level approach algorithm approach. This study aims get best model using pap smear dataset that combined levels with an algorithmic solve imbalanced. The laboratory mostly have few imbalance. Almost every case, minor entities important needed. Over-sampling as a used synthetic minority oversampling technique-nominal (SMOTE-N) adaptive synthetic-nominal (ADASYN-N) algorithms. ensemble classifier AdaBoost bagging classification regression tree (CART) learner-based. obtained from experimental results accuracy, precision, recall, f-measure ADASYN-N AdaBoost-CART.</span>