Heart-Disease: Cleveland Heart Disease Patient Records

The Cleveland Heart Disease database contains 76 attributes per patient, though published experiments typically use a subset of 14. It was created by a consortium including the Hungarian Institute of Cardiology and University Hospitals in Zurich and Basel. The target field indicates the presence of heart disease on an integer scale from 0 (no presence) to 4.

Use Cases

• Binary classification of heart disease presence based on the 'goal' field.
• Multi-class severity prediction using the full 0-4 integer scale of the target.
• Identifying trends in cardiovascular health from the 76 available attributes.
• Benchmarking diagnostic models against the widely used 14-feature subset.

Strengths

• Contains 76 attributes per patient record, providing a broad set of potential predictors.
• Target variable is explicitly defined on a 0-4 integer scale for presence/severity.
• The Cleveland subset is a well-established benchmark in machine learning research.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

Hungarian Institute of Cardiology; University Hospitals in Zurich, Basel; V.A. Medical Center, Long Beach; Cleveland Clinic Foundation.

Collection Method

Clinical patient data collection.

Geography

Likely includes data from Budapest, Zurich, Basel, Long Beach, and Cleveland.