Improving Residential Energy Efficiency using Machine Learning: A Predictive Approach to Smart Home Energy Optimization

Nalakath Assi, Mohammed Aslam (2025) Improving Residential Energy Efficiency using Machine Learning: A Predictive Approach to Smart Home Energy Optimization. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (3MB) | Preview
Preview	PDF (Configuration Manual) Download (1MB) | Preview

Abstract

Residential buildings are a large contributor of energy consumption and emissions, thus efficiency improvements are a milestone. The paper identifies and develops implications of interpretable machine learning, classifying and predicting residential energy efficiency based on the U.S. dataset of NREL ResStock. Aspects included structural as well as insulation, mechanical as well as socioeconomic features. The data preprocessing involved the removal of columns, imputing missing values, encoding categorical data and scaling the features, where labels were Efficient, Moderate or Inefficient. The 80/20 split was used to test three classification and three regression models. Random Forest models produced the best results, 87% accuracy (F1 = 0.88), and 0.94 R2 outperforming other comparable research. SHAP analysis showed that the key predictors were CO2 emissions, square footage, and the type of the heating and cooling system, allowing guided recommendations on retrofits. The findings indicate that balanced datasets, variety of features, and ensembles achieves accurate, transparent energy assessment prediction models. Limitations involve cybernetics dependency on simulated data and lack of the complete retrofit module which is found to be improved in coming work at combination with model optimisation and adjustment to diverse housing stocks.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Onwuegbuche, Faithful UNSPECIFIED
Subjects:	H Social Sciences > HC Economic History and Conditions > Natural resources > Power resources > Energy consumption H Social Sciences > HG Finance > Fintech T Technology > T Technology (General) > Information Technology > Fintech Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in FinTech
Depositing User:	Ciara O'Brien
Date Deposited:	24 Jun 2026 10:41
Last Modified:	24 Jun 2026 10:41
URI:	https://norma.ncirl.ie/id/eprint/9393

Actions (login required)

View Item