Breast Cancer Classification
ML model for breast cancer diagnosis with 96.5% accuracy
Overview
Developed a machine learning model to classify breast cancer patients into malignant and benign categories, achieving high accuracy in early detection.
Technical Details
- Technologies: Python, scikit-learn, Keras
- Best Model: Random Forest Classifier
- Accuracy: 96.50%
- Recall: 95.47%
Approach
- Data Preprocessing: Handled missing values and normalized features
- Feature Engineering: Selected most relevant diagnostic features
- Model Selection: Compared multiple classifiers (SVM, Random Forest, Neural Networks)
- Evaluation: Used cross-validation and confusion matrix analysis
Results
The Random Forest classifier achieved the best performance with:
- High accuracy (96.50%) for reliable predictions
- Strong recall (95.47%) to minimize false negatives, crucial for cancer detection
Impact
This project demonstrates the potential of ML in healthcare for early cancer detection, potentially improving patient outcomes through timely diagnosis.