Data & ML
Clinical Data Machine Learning & Survival Modeling
Historical clinical data analysis using machine learning classification models to evaluate post-surgery patient survival probabilities.
System Implementation & Approach
Applied scientific computing and exploratory data analysis (EDA): feature preprocessing, correlation matrix analysis, and supervised machine learning classification modeling on the Haberman Survival dataset.
Role & Contribution
Collaborative analysis: contributed to data cleaning pipelines, feature correlation visualization, and quantitative model evaluation metrics in Python.
Key Implementation Highlights
- Structured exploratory data analysis checking class imbalance across post-surgery age, operation year, and axillary node counts.
- Evaluated correlation distributions and feature normalization prior to supervised classifier fitting.
- Benchmarked classifier predictions with reproducible evaluation metrics in versioned notebooks.
Verification & Outcomes
- Reproducible end-to-end preprocessing, EDA visualization, and classification pipeline verified in Jupyter.