Comparative Analysis of Machine Learning Models in Predictive Analytics for Residential Energy Consumption

Hongyuan Jia

2023

Abstract

Today, household energy consumption patterns are crucial for a sustainable living environment. Many researchers have made contributions in this field. Some researchers use various machine learning models to predict household appliance energy consumption, such as long short-term memory (LSTM) deep neural network, Naive Bayes, and Support Vector Machines. Still other researchers try to reduce electricity usage by maximizing the use of solar power in homes. This study sets out to provide an in-depth analysis of household energy usage patterns, focusing specifically on appliance energy consumption over approximately 4.5 months. Careful analysis was conducted using a dataset merged with meteorological data from an in-home wireless sensor network and the nearest airport weather station. In this research, we explored an array of regression techniques: linear, Ridge, KNN, decision tree, and random forest. By testing multiple models, random forest regression was the more suitable model for this data set because of its better performance. This study seeks to enrich the growing domain of sustainable living and energy management, emphasizing enhanced energy efficiency within residential settings.

Paper Citation

in Harvard Style

Jia H. (2023). Comparative Analysis of Machine Learning Models in Predictive Analytics for Residential Energy Consumption. In Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML; ISBN 978-989-758-705-4, SciTePress, pages 251-255. DOI: 10.5220/0012800500003885

in Bibtex Style

@conference{daml23,
author={Hongyuan Jia},
title={Comparative Analysis of Machine Learning Models in Predictive Analytics for Residential Energy Consumption},
booktitle={Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML},
year={2023},
pages={251-255},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012800500003885},
isbn={978-989-758-705-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML
TI - Comparative Analysis of Machine Learning Models in Predictive Analytics for Residential Energy Consumption
SN - 978-989-758-705-4
AU - Jia H.
PY - 2023
SP - 251
EP - 255
DO - 10.5220/0012800500003885
PB - SciTePress