TY - CHAP
T1 - Machine Learning Algorithms for Energy Consumption Prediction in Smart Homes
T2 - A Comparative Study
AU - Moulla, Donatien Koulla
AU - Attipoe, David
AU - Akinyemi, Lateef Adesola
AU - Thottempudi, Sree Ganesh
AU - Agbo-Ajala, Jelil Olatunbosun
AU - Mnkandla, Ernest
AU - Abran, Alain
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.
PY - 2025
Y1 - 2025
N2 - Accurate energy consumption prediction is important for optimizing energy usage, reducing costs, and minimizing environmental impact, particularly in smart homes. This study presents a comprehensive analysis of six leading machine learning algorithms for energy consumption prediction: long short-term memory (LSTM) networks, random forest (RF), extreme gradient boosting (XGBoost), gated recurrent unit (GRU), support vector machines (SVMs), and artificial neural networks (ANNs). The performance of each model was evaluated using metrics such as the mean squared error (MSE), mean absolute error (MAE), mean absolute percentage error (MAPE), and coefficient of determination (R2). The findings indicated that the RF model consistently outperformed the other models in terms of prediction accuracy, showing the lowest MAE and MSE, and the highest R2, both with and without incorporating weather data. Through examining their performance, strengths, and weaknesses, this study provides valuable insights into the suitability of these algorithms for smart home energy management applications.
AB - Accurate energy consumption prediction is important for optimizing energy usage, reducing costs, and minimizing environmental impact, particularly in smart homes. This study presents a comprehensive analysis of six leading machine learning algorithms for energy consumption prediction: long short-term memory (LSTM) networks, random forest (RF), extreme gradient boosting (XGBoost), gated recurrent unit (GRU), support vector machines (SVMs), and artificial neural networks (ANNs). The performance of each model was evaluated using metrics such as the mean squared error (MSE), mean absolute error (MAE), mean absolute percentage error (MAPE), and coefficient of determination (R2). The findings indicated that the RF model consistently outperformed the other models in terms of prediction accuracy, showing the lowest MAE and MSE, and the highest R2, both with and without incorporating weather data. Through examining their performance, strengths, and weaknesses, this study provides valuable insights into the suitability of these algorithms for smart home energy management applications.
KW - Energy consumption
KW - Internet of Things
KW - Machine learning
KW - Prediction ML models
KW - Smart homes
KW - Time-series forecasting
UR - https://www.scopus.com/pages/publications/105005412899
U2 - 10.1007/978-3-031-80817-3_5
DO - 10.1007/978-3-031-80817-3_5
M3 - Book Chapter
AN - SCOPUS:105005412899
T3 - Lecture Notes on Data Engineering and Communications Technologies
SP - 75
EP - 94
BT - Lecture Notes on Data Engineering and Communications Technologies
PB - Springer Science and Business Media Deutschland GmbH
ER -