Critical Asset Lifecycle Optimisation: Predictive Resilience in National Infrastructure

Comparative analysis of machine learning classification models for predicting water point functionality in Tanzania using the Taarifa dataset. This technical report investigates four classification approaches (K-Nearest Neighbors, Naive Bayes, Random Forest, Decision Tree) with different preprocessing and feature engineering methodologies. Random Forest achieved 92% accuracy using target-based encoding. 

This work was completed as part of MSc Computer Science (AI) research at the University of Nottingham.