Evaluating Machine Learning Models for House Price Prediction

This project explored the use of machine learning to predict house prices in California using data from housing and crime statistics. Five models were developed and evaluated: Linear Regression, Random Forest, XGBoost, LightGBM and K-Nearest Neighbours. 

View report >>

Driver Injury Severity Prediction

I conducted an in-depth analysis of traffic collision data to predict the severity of driver injuries using machine learning techniques. The work involved extensive data preprocessing, exploratory data analysis and feature selection to identify key factors associated with injury outcomes. I developed and evaluated several models, including logistic regression and decision trees, applying hyperparameter tuning and cross-validation to optimise performance. 

View report >>

View Python notebooks >>