BrianRono7/kenya-breast-cancer-ml

Machine Learning & Deep Learning research on breast cancer diagnosis, prognosis, and treatment safety. This solution is tailored for Kenya’s healthcare needs. # Breast Cancer Classification Project ## Overview This project builds and evaluates machine-learning models to classify breast tumors as **Malignant (M)** or **Benign (B)** using numerical clinical features. The workflow includes data exploration, feature analysis, modeling, and evaluation. --- ## Dataset The dataset contains: - **Diagnosis** (target) - Numerical features describing tumor characteristics: - Radius - Texture - Perimeter - Area - Smoothness - Compactness - Concavity - Symmetry - Fractal Dimension Each feature includes mean, standard error, and worst values. --- ## Workflow 1. Load and inspect the dataset 2. Exploratory Data Analysis (EDA) 3. Correlation and multicollinearity checks 4. Preprocessing and feature engineering 5. Train models: - Logistic Regression - Random Forest - K-Nearest Neighbors - AdaBoost - XGBoost 6. Evaluate models using: - Accuracy - Confusion Matrix - Classification Report --- ## How to Run Install dependencies: ```bash pip install numpy pandas matplotlib seaborn scikit-learn xgboost