Implementasi Data Mining Dengan Algoritma Multiple Linear Regression Untuk Memprediksi Penyakit Diabetes

Abstract

According to WHO, diabetes is a metabolic disorder characterized by high levels of sugar in the blood. Diabetes is a deadly disease if the sufferer cannot control it and it will become a complication. Many people are affected by diabetes and find out too late, so that at the time of treatment the condition has complications. Early detection of diabetes is very helpful for sufferers to avoid complications that will occur. Therefore we need a data mining technique that can process data and prevent diabetes from an early age. Data mining is a process of extracting knowledge from a number of data to find a pattern. Data mining has been widely used, one of which is the prediction method to find out people with diabetes. There are so many prediction methods available, one of which is linear regression, where this method uses dependent and independent attributes. In this study, the multiple linear regression method is used to predict diabetes, and evaluates using RMSE (root mean square error). The results of this study produce an RMSE value of 0.403, the RMSE test uses cross validation by changing the number of validation value