Data Analysis and Prediction of Diabetic Disease by Supporting Data Mining

Abstract

Diabetes is the world's prevalent and fast-growing illnesses. For all nations, it is the greatest health problem. Diabetes is regarded one of the deadliest and most chronic diseases causing blood sugar to rise. Diabetes is regarded one of the deadliest and most chronic diseases causing blood sugar to rise. If diabetes remains untreated and unidentified, there are many complications. The tedious process is defining outcomes in a patient's visit to a diagnostic centre and doctor's advice. The increase in approaches to machine learning, however, solves this critical issue. This study's motivation is to develop a model that can predict the probability of peak precision in patients with diabetes. For this research, we use data mining tools that WEKA. WEKA is machine-learning software. Weka has its working procedure. Machine learning classification algorithms; are used in this experiment to identify diabetes at an early point, namely Decision Tree, logistic function and Naive Bays. All three algorithms ' performances were assessed on multiple measures such as Precision, Accuracy, F-Measure, and Recall. Accuracy measured over cases classified properly and wrongly. Results achieved shows outperform of logistical feature with the greatest precision of 78.26 percent compared to other algorithms. These findings verified correctly and systematically using Receiver Operating Characteristic (ROC) curves. At the end the predication of each are attribute are registered.