Diabetes Diagnosis using Ensemble Models in Machine Learning

Abstract

Diabetes is one of the diseases where early detection is must given the fact that it is not possible to cure the disease once the patient gets the diabetes disease. As the number of patients is increasing on a day to day basis, it is difficult for the doctors to perform manual detection. With the technologies like Machine Learning in hand, we can perform automative detection to some extent. Lot of research has been performed till now on this diabetes diagnosis problem. This paper discusses predictive analysis using two ensemble machine Learning Algorithms such as Random Forest and GBDT. In this paper, we have performed various Experiments on Pima Indians Diabetes Dataset which contains Diabetes patients record and results are discussed. This paper additionally discusses the importance of Interpretability of output in the healthcare domain and explains how it will help the doctors in real time if we could provide interpretability of the output along with the output of the patient record given by machine learning model.