PENERAPAN SYNTHETIC MINORITY OVERSAMPLING TECHNIQUE (SMOTE) TERHADAP DATA TIDAK SEIMBANG PADA PEMBUATAN MODEL KOMPOSISI JAMU

Abstract

As the times many people use herbal remedies(jamu) to address health issues. Herbal medicines are madefrom plants with a specific composition to produce certainproperties, so a model is needed to be made in order tofind the right formula to make herbal medicine with certainproperties. In this study, the response being investigated is apotent herbal medicine in treating mood and behavior disorder.In this analysis, the model is developed using logistic regression.The accuracy of the model can be seen from the Area UnderCurve (AUC). Imbalanced data on the response variable cancause the value of AUC become low. One of the ways tosolve it is using Synthetic Minority Oversampling Technique(SMOTE). From this analysis, Nagelkerke R2 values generatedby the model with SMOTE 3.2% lower than model withoutSMOTE. Nonetheless, the model with SMOTE is more accuratethan model without SMOTE because has higher AUC value.The resulting AUC is equal to 0.976 for the model with SMOTEand 0.908 for model without SMOTE. The results show thatSMOTE can increase the accuracy of the model for imbalanceddata.Keywords-imbalance data, logistic regression, SMOTE