KLASIFIKASI DIABETES PADA WANITA MENGGUNAKAN METODE NAIVE BAYES CLASSIFIER

Abstract

The report from Riskesdas shows that there is a 2x increase in diabetes every year in Indonesia. This is due to an increase in factors such as human population, age, obesity, irregular eating patterns and lack of physical activity. The increase in a factor that causes diabetes in Indonesia must be prevented. The first step in preventing diabetes is to detect the risk factors for diabetes that may occur. Influencing factors include behavioral factors and sociodemographic factors The increase in diabetes in a country is due to late identified factors. The number of factors that are collected in order to detect whether a person has diabetes or not requires a fairly large data processing system. The data used in this study are diabetes data obtained from the Pima Indian Diabetes Database with attributes of pregnant, glucose, diastolic, triceps, insulin, BMI, history of diabetes, age and 300 data output. The Naive Bayes Classifier method can be used to classify diabetes in women based on pregnant, glucose, diastolic, triceps, insulin, BMI, history of diabetes, age and output. The accuracy result of the Naive Bayes Classifier method in classifying diabetes in women is 84% ​​of 300 data which is divided into 2, namely 275 data as training data and 25 data as test data.