Pima Indian Diabetes Dataset for Diagnostic Prediction

The Diabetes-Data-Set originates from the National Institute of Diabetes and Digestive and Kidney Diseases. It contains diagnostic measurements from female patients of Pima Indian heritage, aged at least 21, to predict the onset of diabetes. The dataset was used in a 1988 study that applied the ADAP learning algorithm to forecast diabetes mellitus.

Use Cases

• Train binary classifiers to predict diabetes diagnosis based on plasma glucose and insulin levels.
• Analyze the correlation between body mass index (BMI) and diabetes outcome in a specific demographic.
• Build interpretable models to understand the influence of age and number of pregnancies on diabetes risk.
• Benchmark new predictive algorithms against the established ADAP algorithm results described in the literature.

Strengths

• Dataset has a clear, medically relevant prediction target (diabetes diagnosis).
• Includes multiple physiological measurements per patient, such as glucose, blood pressure, and skin thickness.
• The population is well-defined, consisting solely of adult females of Pima Indian heritage.
• Has documented past usage in published academic research, providing a benchmark.

Limitations

• Row count and dataset size are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Data reflects a specific geographic and demographic bias, focusing on a Pima Indian community near Phoenix, Arizona.

Provenance

Source

National Institute of Diabetes and Digestive and Kidney Diseases

Collection Method

Diagnostic measurements collected, likely from medical surveys and examinations.

Geography

Population lives near Phoenix, Arizona, USA.