Diabetes Forecasting Using Supervised Learning Techniques

Abstract

Diabetes Mellitus is one of the most serious health challenges affecting children, adolescents and young adults in both developing and developed countries. To predict hidden patterns of diseases diagnostic in the healthcare sector, nowadays we use various data mining techniques. In this paper, we have applied supervised machine learning techniques like Naive Bayes and J48 decision tree to identify diabetic patients. We evaluated the proposed methods on Pima Indian diabetes data sets, which is a data mining data sets from UCI machine learning laboratory. It has been observed through analysis of the experimental results that Naive Bayes performs better than the decision tree method J48.