Katie Wood Email & Phone Number

Graduated with a MSc Computer Science degree at The University of York with a distinction grade.A passion for data science, machine learning, and software development to solve real-world problems. My master's research project earned a high distinction with an 81% score. It systematically analysed the precision of Random Forest and XGBoost in predicting university student dropout, for early detection of vulnerable students and providing useful insights for educational institutions and decision makers.The research explored:- Model Suitability and Comparison: Investigated the effectiveness of Random Forest and XGBoost, with comparisons to Logistic Regression to determine the best model for the task - Model Interpretation with SHAP: Used Shapley Additive Explanations (SHAP) to interpret model decisions, giving valuable insights into the key risk factors that contribute to student dropout.- Evaluation Metrics: Accuracy, precision, recall, and F-measure, plotting Precision-Recall (PR) curves and ROC curves.- Hyperparameter Tuning: Optimized model performance through grid search and 3-fold cross-validation.Following a standard machine learning pipeline, including, data exploration, preprocessing, train-test split, hyperparameter tuning, model building, and detailed analysis.