Welcome to my Data Science Projects Repository! This collection showcases my expertise in machine learning, predictive modeling, and statistical analysis. Each project demonstrates end-to-end ML workflows—from data preprocessing and feature engineering to model deployment and evaluation.
🧠 Model Performance Focus – I prioritize accuracy, interpretability, and real-world applicability
🔬 Rigorous Experimentation – Multiple algorithms tested, hyperparameters tuned, results validated
📈 Business-Driven ML – Every model solves a concrete problem with measurable outcomes
⚙️ Production-Ready Code – Clean, reproducible workflows with proper documentation
Built binary classification system to predict cancer diagnosis (0 = benign, 1 = malignant) using patients' medical history. Compared three algorithms with rigorous cross-validation and hyperparameter tuning.
Impact: XGBoost achieved 94.3% accuracy, providing reliable early detection support
Models Tested: Logistic Regression (86%) • Random Forest (92.3%) • XGBoost (94.3%)
Stack: 
Developed churn prediction model analyzing demographic, contract, and billing patterns to identify at-risk customers. Enabled proactive retention strategies for telecom company.
Impact: SVM model achieved 83% accuracy, identifying churn drivers for targeted interventions
Models Tested: Naive Bayes (75%) • Logistic Regression (82%) • SVM (83%)
Stack:
Built regression model to estimate property values in Bandung using features like land area, building size, and structural attributes. Applied advanced feature engineering and ensemble methods.
Impact: XGBoost explained 78.38% of price variance, outperforming traditional regression approaches
Models Tested: Linear Regression (72.13%) • Random Forest (77.65%) • XGBoost (78.38%)
Stack:
Machine Learning: Classification • Regression • Ensemble Methods • Model Evaluation
Libraries & Frameworks: Scikit-learn • XGBoost • Pandas • NumPy • Matplotlib • Seaborn
Core Competencies: Feature Engineering • Hyperparameter Tuning • Cross-Validation • Model Interpretability