Predictive and Comparative Analysis for Diabetes using Machine Learning Algorithms
Abstract
Diabetes is a condition which is detected when the blood glucose levels in the body is high due to the lack of presence of Insulin. Increased sugar levels in blood leads to complications like heart diseases and stroke. With increased use of Machine Learning techniques in the healthcare sectors, it becomes particularly important to detect diabetes in the early stages. This research implements machines learning algorithms like Logistic Regression, Decision Tree, Random Forest, Neural network and K-nearest neighbors to predict diabetes. The most essential features for prediction was observed and the optimum method for predictive analysis of diabetes was obtained for the dataset.