Summary#

Experienced with scikit-learn for training, evaluating, and tuning machine learning models. Applied to classification, regression, and feature selection tasks.

How I Apply This Skill#

• Trained Random Forest classifiers achieving 83.8% accuracy on Titanic data
• Applied Boruta feature selection for identifying important predictors
• Implemented train/test splits and cross-validation for model evaluation
• Used preprocessing pipelines for encoding, scaling, and data transformations
• Evaluated models with accuracy, AUC, precision/recall metrics

Key Strengths#

• Model Training: Random Forest, Decision Trees, SVM, Naive Bayes, Neural Networks
• Feature Selection: feature importance, correlation-based
• Evaluation: Cross-validation, confusion matrices
• Preprocessing: StandardScaler, LabelEncoder, train_test_split

Related Projects#

• Long COVID Prediction
• Titanic Survival