A Python project that learns spending patterns from 180 days of synthetic expense data. It compares a Random Forest, linear regression and gradient boosting against a baseline using cross-validation, flags unusual expenses with an Isolation Forest, and runs 1,000 Monte Carlo simulations to estimate the risk of going over budget. It includes an interactive dashboard and a README that explains its limits. Tech: Python, pandas, scikit-learn.

A machine learning project that predicts which language-centre enquiries will enrol, using 600 synthetic records (source, reply time, follow-ups, trial class). Logistic regression and Random Forest were compared with a baseline using 5-fold cross-validation, and both beat it (AUC 0.74 and 0.71 vs 0.50). Tech: Python, pandas, scikit-learn.