Combining Advanced Feature-Selection Methods to Uncover Atypical Energy-Consumption Patterns

Lucas Henriques (), Felipe Prata Lima and Cecilia Castro ()
Additional contact information
Lucas Henriques: Centre of Mathematics, Universidade do Minho, 4710-057 Braga, Portugal
Felipe Prata Lima: IT Department, Federal Institute of Alagoas, Maceió 57020-600, AL, Brazil
Cecilia Castro: Centre of Mathematics, Universidade do Minho, 4710-057 Braga, Portugal

Future Internet, 2024, vol. 16, issue 7, 1-23

Abstract: Understanding household energy-consumption patterns is essential for developing effective energy-conservation strategies. This study aims to identify ‘out-profiled’ consumers—households that exhibit atypical energy-usage behaviors—by applying four distinct feature-selection methodologies. Specifically, we utilized the chi-square independence test to assess feature independence, recursive feature elimination with multinomial logistic regression (RFE-MLR) to identify optimal feature subsets, random forest (RF) to determine feature importance, and a combined fuzzy rough feature selection with fuzzy rough nearest neighbors (FRFS-FRNN) for handling uncertainty and imprecision in data. These methods were applied to a dataset based on a survey of 383 households in Brazil, capturing various factors such as household size, income levels, geographical location, and appliance usage. Our analysis revealed that key features such as the number of people in the household, heating and air conditioning usage, and income levels significantly influence energy consumption. The novelty of our work lies in the comprehensive application of these advanced feature-selection techniques to identify atypical consumption patterns in a specific regional context. The results showed that households without heating and air conditioning equipment in medium- or high-consumption profiles, and those with lower- or medium-income levels in medium- or high-consumption profiles, were considered out-profiled. These findings provide actionable insights for energy providers and policymakers, enabling the design of targeted energy-conservation strategies. This study demonstrates the importance of tailored approaches in promoting sustainable energy consumption and highlights notable deviations in energy-use patterns, offering a foundation for future research and policy development.

Keywords: behavior analysis; consumption patterns; feature selection; fuzzy rough sets; random forest (search for similar items in EconPapers)
JEL-codes: O3 (search for similar items in EconPapers)
Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
https://www.mdpi.com/1999-5903/16/7/229/pdf (application/pdf)
https://www.mdpi.com/1999-5903/16/7/229/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jftint:v:16:y:2024:i:7:p:229-:d:1424545

Access Statistics for this article

Future Internet is currently edited by Ms. Grace You

More articles in Future Internet from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().