<![CDATA[Data IDF states that there will be 19.47 million people with diabetes in Indonesia in 2021. One of the causes is a lack of awareness of healthy food consumption which has an impact on increasing body weight due to high blood sugar. In overcoming this incident, it is necessary to implement an application to detect diabetes. The use of an expert system provides many conveniences for someone in health examinations and also for health workers or doctors in diagnosing a disease. In diagnosing, an expert system requires a method, one of which is the Naive Bayes classification method. In this study, data on diabetes sufferers was taken from the Kaggle site as training data, with a total of 768 data from nine attributes and one of the attributes as a label. To make it easier to operate the application, only four attributes were used as test samples based on calculating the highest correlation value, namely pregnancy, glucose, BMI and age. Next, the opinion of a health expert, namely a specialist in internal medicine, and calculating the accuracy value of diabetes data using the Naïve Bayes algorithm. The resulting data accuracy value is 79%. Implementation of a web-based expert system application using the PHP programming language and MySQL database. This expert system application aims to detect diabetes based on the results of a health check to predict whether it is positive for diabetes or negative for diabetes.]]>
Copyrights © 2025
