Heart Disease Classification
A supervised-learning notebook that compares classification approaches on clinical attributes from the Cleveland heart disease dataset.
Problem
Can clinical attributes in the Cleveland Heart Disease dataset support a basic binary classification experiment?
The notebook is the primary artifact. It covers data preparation, feature handling, and model training with scikit-learn. The experiment is educational and does not provide a clinical decision tool.
Problem Statement
Given clinical parameters about a patient, can we predict whether or not they have heart disease?
Evaluation boundary
Accuracy is used as a project target in the notebook. It should not be treated as clinical performance without class-specific metrics, validation design, calibration, external data, and clinical review.
Product value and limitations
The notebook is useful for understanding a complete tabular-classification workflow: define the target, prepare clinical attributes, compare models, and inspect the evaluation. The dataset and notebook do not establish generalization to current patients or other clinical settings. Predictions must not replace medical assessment.
Interactive Notebook
Explore the full data analysis, feature engineering, and model training process below.
Loading notebook...