Classification of Pregnant Women's Risk Using the Naïve Bayes Ensemble Method with Bagging Technique

Abstract

The risk of pregnant women is the potential for complications or health problems that can affect the mother and fetus during pregnancy and childbirth. This study aims to determine the classification of the risk of pregnant women using the Naïve Bayes method with the Ensamble Bagging technique and also to determine the performance of naïve bayes with the bagging technique in classifying pregnant women who are at risk and not at risk. The data of pregnant women used is data taken from the Panambugan Health Center which is then Pre-processing data to obtain ideal data conditions for processing. The entire data set will be divided into training data and test data into 9 data partitions. The results obtained show that the Naïve Bayes method with the Ensamble Bagging technique using the help of Oversampling SMOTE to increase data imbalance gets an increase in Accuracy 8%, Precision 4%, Recall 8%, and F1-Score 8% higher compared to Naïve Bayes.