A machine learning project focused on dermatological disease classification using ensemble learning algorithms.
The project compares Random Forest and Gradient Boosting models using performance metrics, cross-validation, confusion matrices, and feature importance analysis.
- Dermatological disease classification
- Random Forest implementation
- Gradient Boosting implementation
- Cross-validation analysis
- Feature importance analysis
- Correlation matrix analysis
- Confusion matrix evaluation
- Overfitting analysis
- Python
- Scikit-learn
- Pandas
- NumPy
- Matplotlib
- Seaborn
- Dermatology Dataset
- UCI Machine Learning Repository
- 366 patient records
- 34 clinical features
- 6 disease classes
- Accuracy
- Precision
- Recall
- F1-score
- AUC
- Cross-validation scores
This project was developed to improve machine learning model evaluation, medical data analysis, and ensemble learning skills.