rifqinvnd/Stroke-Disease-Classification

Detection (Prediction) of the possibility of a stroke in a person

20
/ 100
Experimental

Implements K-Nearest Neighbors classification with SMOTE-based resampling to handle severe class imbalance in stroke detection across 11 clinical features (age, glucose levels, BMI, hypertension, heart disease, smoking status). Uses HalvingGridSearchCV for hyperparameter optimization prioritizing recall over accuracy, achieving 97.7% recall to maximize true positive identification of at-risk patients. Built with scikit-learn on Kaggle's 5,110-sample stroke prediction dataset, preprocessed via mean imputation, IQR-based outlier removal, categorical encoding, and StandardScaler normalization.

No commits in the last 6 months.

No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 1 / 25
Community 14 / 25

How are scores calculated?

Stars

9

Forks

3

Language

Jupyter Notebook

License

Last pushed

Jul 09, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/rifqinvnd/Stroke-Disease-Classification"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.