Prediction of Diabetes Mellitus Using Classification Techniques

Sirsi Chandra Shekar and N.V. Sailaja

Diabetes Mellitus is a progressive condition which is rapidly widespread, characterised by the failure of the body to metabolise glucose. The goal of this research was to create an efficient predictive model based on patient data and laboratory findings during their visits to medical facilities with high predictive precision. Using the most recent reports, along with their laboratory details, of 768 Pima Indian Diabetes Database patients aged between 18 and 90 years. Using classification methods, we developed a statistical model. We compared these classification approaches such as Decision Tree (DT), K Nearest Neighbour (KNN), Support Vector Machine (SVM), Random Forest (RF) models and found that the Random Forest performs better than other classification models, providing a 92.30 percent higher accuracy.

Volume 12 | Issue 9

Pages: 46-51

DOI: 10.5373/JARDCS/V12I9/20202640