Health and Lifestyle Biomarker Dataset for Diabetes Risk Probability Estimation and Health Status Classification

This dataset contains 500 structured health records* collected for metabolic health analysis and diabetes risk probability estimation. It includes 6 columns, consisting of 5 numerical features, sugar_level, bp_level, fat_level, fiber, and calories_burn and 1 categorical target variable, health_status. The numerical variables represent blood glucose level, blood pressure, body fat percentage, daily dietary fiber intake, and estimated calories burned, respectively. The target label health_status is a multi-class categorical variable with three classes: **Good, Moderate, and Poor*, enabling supervised machine learning and classification modeling. The dataset is suitable for descriptive statistical analysis, correlation studies, feature importance evaluation, and predictive modeling to estimate diabetes risk probability. It is formatted as a structured tabular dataset and is intended for research, educational, and healthcare analytics applications.