Health Condition Forecaster Using Machine Learning

Abstract

The growing burden of chronic diseases highlights the need for early and reliable prediction. This study presents a machine learning framework, the Health Condition Forecaster, to estimate risks for seven conditions: diabetes, heart disease, Parkinson’s disease, hypertension, stroke, liver disease, and lung cancer. Using cleaned and balanced clinical datasets, models in particular Tree Ensemble Model, Classification Tree, Logistic Regression, Naive Bayes, and K-Nearest Neighbors were evaluated. The findings indicate that achieved higher accuracy and consistency. The system, deployed through a simple clinical interface, demonstrates potential to support timely diagnosis and strengthen preventive healthcare.