<![CDATA[Diabetes is a chronic disease caused by impaired glucose metabolism in the body. In type 1 diabetes, the body's immune system attacks and destroys the insulin-producing cells in the pancreas, so that the body is unable to produce sufficient amounts of insulin. In type 2 diabetes, the body is still able to produce insulin, but is unable to use it effectively. Insulin is a hormone that is useful for controlling glucose in blood cells. If glucose in the blood is not controlled it can cause a number of problems, such as heart disease, stroke, blindness, nerve damage and so on. So the research in this study formulated an increase in accuracy in the diabetes detection process using the Deep Neural Network (DNN) method which was enhanced with the chi-square and PSO methods through the attribute selection process. The results of testing the PIMA dataset with DNN obtained an accuracy value of 76.62% with an AUC value of 0.772. Meanwhile, testing using the DNN method for attribute selection using the Chi-square method and optimization with PSO obtained an accuracy value of 85.71% with an AUC value of 0.818. So it can be concluded from testing diabetes data using the Deep Neural Network method which adds Chi-square as a selection attribute and is optimized using PSO which is better when compared to the DNN method.]]>
Copyrights © 2024
