On the feature engineering of building energy data mining

Understanding the underlying dynamics of building energy consumption is the very first step towards energy saving in building sector; as a powerful tool for knowledge discovery, data mining is being applied to this domain more and more frequently. However, most of previous researchers focus on model...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhang, Chuan, Cao, Liwei, Romagnoli, Alessandro
Other Authors: School of Mechanical and Aerospace Engineering
Format: Article
Language:English
Published: 2020
Subjects:
Online Access:https://hdl.handle.net/10356/139586
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-139586
record_format dspace
spelling sg-ntu-dr.10356-1395862020-05-20T07:11:25Z On the feature engineering of building energy data mining Zhang, Chuan Cao, Liwei Romagnoli, Alessandro School of Mechanical and Aerospace Engineering Engineering::Mechanical engineering Building Energy Feature Engineering Understanding the underlying dynamics of building energy consumption is the very first step towards energy saving in building sector; as a powerful tool for knowledge discovery, data mining is being applied to this domain more and more frequently. However, most of previous researchers focus on model development during the pipeline of data mining, with feature engineering simply being overlooked. To fill this gap, three different feature engineering approaches, namely exploratory data analysis (EDA) as a feature visualization method, random forest (RF) as a feature selection method and principal component analysis (PCA) as a feature extraction method, are investigated in the paper. These feature engineering methods are tested with a building energy consumption dataset with 124 features, which describe the building physics, weather condition, and occupant behavior. The 124 features are analyzed and ranked in this paper. It is found that although feature importance depends on specific machine learning model, yet certain features will always dominate the feature space. The outcome of this study favors the usage of effective yet computationally cheap feature engineering methods such as EDA; for other building energy data mining problems, the method proposed in this study still holds important implications since it provides a starting point where efficient feature engineering and machine learning models could be further developed. NRF (Natl Research Foundation, S’pore) 2020-05-20T07:11:25Z 2020-05-20T07:11:25Z 2018 Journal Article Zhang, C., Cao, L., & Romagnoli, A. (2018). On the feature engineering of building energy data mining. Sustainable Cities and Society, 39, 508-518. doi:10.1016/j.scs.2018.02.016 2210-6707 https://hdl.handle.net/10356/139586 10.1016/j.scs.2018.02.016 2-s2.0-85044618869 39 508 518 en Sustainable Cities and Society © 2018 Elsevier Ltd. All rights reserved.
institution Nanyang Technological University
building NTU Library
country Singapore
collection DR-NTU
language English
topic Engineering::Mechanical engineering
Building Energy
Feature Engineering
spellingShingle Engineering::Mechanical engineering
Building Energy
Feature Engineering
Zhang, Chuan
Cao, Liwei
Romagnoli, Alessandro
On the feature engineering of building energy data mining
description Understanding the underlying dynamics of building energy consumption is the very first step towards energy saving in building sector; as a powerful tool for knowledge discovery, data mining is being applied to this domain more and more frequently. However, most of previous researchers focus on model development during the pipeline of data mining, with feature engineering simply being overlooked. To fill this gap, three different feature engineering approaches, namely exploratory data analysis (EDA) as a feature visualization method, random forest (RF) as a feature selection method and principal component analysis (PCA) as a feature extraction method, are investigated in the paper. These feature engineering methods are tested with a building energy consumption dataset with 124 features, which describe the building physics, weather condition, and occupant behavior. The 124 features are analyzed and ranked in this paper. It is found that although feature importance depends on specific machine learning model, yet certain features will always dominate the feature space. The outcome of this study favors the usage of effective yet computationally cheap feature engineering methods such as EDA; for other building energy data mining problems, the method proposed in this study still holds important implications since it provides a starting point where efficient feature engineering and machine learning models could be further developed.
author2 School of Mechanical and Aerospace Engineering
author_facet School of Mechanical and Aerospace Engineering
Zhang, Chuan
Cao, Liwei
Romagnoli, Alessandro
format Article
author Zhang, Chuan
Cao, Liwei
Romagnoli, Alessandro
author_sort Zhang, Chuan
title On the feature engineering of building energy data mining
title_short On the feature engineering of building energy data mining
title_full On the feature engineering of building energy data mining
title_fullStr On the feature engineering of building energy data mining
title_full_unstemmed On the feature engineering of building energy data mining
title_sort on the feature engineering of building energy data mining
publishDate 2020
url https://hdl.handle.net/10356/139586
_version_ 1681058684970467328