A collection of selected Kaggle projects demonstrating experience with machine learning, statistical modeling, predictive analytics, and model evaluation.
Kaggle Playground Series — Season 5, Episode 3
Built a classification pipeline to predict rainfall using multiple machine-learning approaches and compared their performance using cross-validation and hyperparameter optimization.
Models evaluated:
- Logistic Regression
- Decision Tree
- K-Nearest Neighbors
- XGBoost
- Multi-Layer Perceptron (MLP)
Final model: MLPClassifier
ROC-AUC: 0.89
- Python
- pandas
- scikit-learn
- XGBoost
- Neural networks
- Classification
- Cross-validation
- Hyperparameter optimization
- Model evaluation
- Exploratory data analysis
These projects represent independent applications of machine-learning and statistical modeling techniques to real-world and competition datasets.