Diabetes Prediction Using Machine Learning

Abstract

Diabetes is a chronic condition with elevated blood sugar levels, leading to severe complications if undiagnosed or untreated. The diagnosis process often involves patient referrals and multiple consultations. Predictive analytics in healthcare offers potential for timely decision-making based on patient data.This study aims to develop an accurate model for predicting diabetes using machine learning. The dataset is split into training, validation, and testing sets, with each set serving a distinct purpose. Various classification algorithms are employed, including Logistic Regression, K Nearest Neighbors, Gaussian Naive Bayes, Support Vector Machines, and more.The study uses the Pima Indians Diabetes Database (PIDD), which contains diagnostic measures for diabetes detection. Performance metrics such as Precision, Accuracy, Specificity, and Recall are calculated using confusion matrix analysis.By comparing algorithm performance, the study identifies the best approach for early diabetes detection. The goal is to assist healthcare professionals in diagnosing diabetes sooner, improving patient outcomes, and reducing complications.