Using machine learning to identify primary features in choosing electric vehicles based on income levels - Data Science and Management 7 - Issue 1

Aug 5, 2026

Upload by Hazel

An electric vehicle is increasingly popular, yet concerns about battery duration limit adoption. This study identifies key factors influencing the likelihood of owning an electric vehicle across household income levels in the U.S. The dataset is split into three income subgroups: $50k–$150k, $150k–$250k, and $250k+. Several supervised machine learning classifiers are evaluated, and Naive Bayes achieves comparatively stronger overall accuracy and F1 scores. Probability analysis shows one-way commuting distance is the most important factor for all income groups.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages7

Reading Time18 min

AccessFree to download

Abstract

Keywords

1. Introduction

Research aim and methodology

Motivation and prior findings

Boston University OpenBU [http://open. bu.edu](http://open. bu.edu) BU Open Access Articles BU Open Access Articles 2024-03 Using machine learning to identify primary features in choosing electric vehicles based on inc. . . This work was made openly accessible by BU Faculty. Please share how this access benefits you. Your story matters. | Version | Published version | | --- | --- | | Citation (published version): | M. Ma, E. Pinsky. 2024. "Using machine learning to identify primary features in choosing electric vehicles based on income levels" Data Science and Management, Volume 7, Issue 1, pp.1-6 . [https://doi.org/10.1016/j.dsm.2023.10.001](https://doi.org/10.1016/j.dsm.2023.10.001) | [https://hdl.handle.net/2144/48814](https://hdl.handle.net/2144/48814)[ ](https://hdl.handle.net/2144/48814)Boston University Data Science and Management 7 (2024) 1–6 Contents lists available at ScienceDirect Data Science and Management journal [homepage:](homepage: www.keaipublishing.com/en/journals/data-science-and-management)[ www.keaipublishing.com/en/journals/data-science-and-management](homepage: www.keaipublishing.com/en/journals/data-science-and-management) Research article Using machine learning to identify primary features in choosing electric vehicles based on income levels * Mingjun Ma , Eugene Pinsky Department of Computer Science, Boston University Metropolitan College, Boston, 02215, USA | A R T I C L E I N F O | | A B S T R A C T | | --- | --- | --- | | Keywords:
Unbabalced data
Electric vehicle
Machine learning
Sampling with replacement
Supervised learning
Naive Bayes | | An electric vehicle is becoming one of the popular choices when choosing a vehicle. People are generally impressed with electric vehicles’ zero-emission and smooth drives, while unstable battery duration keeps people away. This study tries to identify the primary factors that affect the likelihood of owning an electric vehicle based on different income levels. We divide the dataset into three subgroups by household income from $50,000 to $150,000 or low-medium income level, $150,000 to $250,000 or medium-high income level, and $250,000 or above, the high-income level. We considered several machine learning classiﬁers, and naive Bayes gave us a relatively higher accuracy than other algorithms in terms of overall accuracy and F1 scores. Based on the probability analysis, we found that for each of these groups, one-way commuting distance is the most important for all three income levels. | 1. Introduction An electric vehicle is one of the popular choices for vehicle owners. Zero emission and a smooth driving experience are the potential reasons for it. According to Majumder (2021), the mechanical efﬁciency of an electric vehicle is between 60% and 70%, while the efﬁciency ofa vehicle with an internal combustion engine is 18%–22%(Majumder, 2021). On the other side, short environmental protection agency (EPA) range and charging issues keep people from electric vehicles as the average range of gasoline-powered vehicles is 300 miles (Bonges and Lusk, 2016). In comparison, the average range of an electric vehicle is 110 miles (Bongesand Lusk, 2016). The battery of an electric vehicle is not capable of supporting long-distance driving without charging stations. However, electric vehicles are popular in some regions regardless of traveling range. A number of other authors (Axsen et al., 2016; Carley et al., 2013; Sovacool et al., 2019) suggested that high annual-income households show more interest and are more likely to have electric vehicles at home than low-income households. The price of an electric vehicle is generally more expensive than other types of vehicles. Furthermore, those earning more than 90,000 euros per year are more concerned about fuel efﬁciency, technical reliability and safety, while households with less than 10,000 euros are concerned about ﬁnancial savings (Sovacool et al., 2019).

FAQs

What is the main goal of the study?

To predict electric-vehicle ownership using classification machine learning models and identify the primary factors driving choices across different U.S. income levels.

How is the dataset divided for the income analysis?

It is split into three household income subgroups: $50,000–$150,000, $150,000–$250,000, and $250,000 or above.

Which machine learning model performs best in the study?

Naive Bayes shows relatively higher overall accuracy and F1 scores compared with other evaluated algorithms.

What factor is found to be most important across income groups?

One-way commuting distance is identified as the most important factor for all three income levels.

Aug 5, 2026

Upload by Hazel