Test Model To Predict Diabetes Using Machine Learning Algorithm

Due to change in lifestyle, Diabetic Mellitus (DM) has become a serious threat in the health-care systems in developed as well as developing countries. The proposed work aims to forecast a person’s likelihood of having diabetes using machine learning approach. The occurrence of diabetic in a person depends on many healthcare parameters like number of pregnancies, insulin level, age, and BMI and many more. In this paper, there has been inference of a binary classification model based on the available historical data then their performances are compared using several machine learning. In the present study dataset is collected from NIDDK Diseases, where every patient is an Indian woman from Pima group who is at least 21 years old. In the examinations, researchers determined that XGBoost has demonstrated both performance and speed prowess. Classifier performs better than other Machine learning algorithms.