Research on Potential Customer Mining Models for Electric Vehicles Based on Machine Learning Models and Service Growth Models
DOI:
https://doi.org/10.54097/xmetzj22Keywords:
Single factor variance, Logistic regression, Service intensity, Service growth modelAbstract
This study investigates the factors that impact the sales of electric vehicles (EVs) across different brands. Initially, multiple linear regression was performed on the data, and the variance inflation factors were all found to be less than 10. For the first category of indicators, one-way ANOVA was conducted to separate the three types of brands. For the second category of indicators, a logistic regression model was developed to complete the selection. Through multiple rounds of filtering, the following indicators were determined: For the first category, indicators affecting Brand 1 are A1, A2, A3, A4; indicators affecting Brand 2 are A1, A3; indicators affecting Brand 3 are A1, A2, A3, A5. For all brands, the second category of influencing indicators includes B4, B10, B11, B16, B17. A potential customer mining model was then established. Logistic regression models were created based on the sales influencing factors for different brands, and the model formulas for the three types of brands were calculated. The purchase intentions of 15 target customers were predicted, with an accuracy rate of up to 90%. Finally, the study explores whether increasing service efforts could change customer purchase intentions. A stepwise percentage service growth model was established, incorporating an initial service difficulty coefficient to determine how service satisfaction could be altered. Customers 2, 8, and 11 were selected for validation, and their respective service increment percentages were derived, leading to sales strategies tailored to these three customers.
References
[1] Wang Shengjie, Zhang Qinghong. Research on customer churn prediction in telecommunication industry based on explainable machine learning model [J]. Telecommunication Science, 2024, 40 (07): 121-133.
[2] WANG Yu-lin. Research on Customer Churn Prediction Model of Broadcasting and Television based on Ensemble Learning [J]. TV technology, 2024, 48 (07) : 59-66. DOI: 10.16280 / j.v ideoe. 2024.07.014.
[3] GAO Tian-Chen, Qu Hao, Wang Feifei, et al. Research on customer churn warning model of securities companies based on high-dimensional feature factors [J]. Journal of Economic Management, 2023, 2 (04): 143-168.
[4] Li Ziyi, Wu Xinyang. Research on expert Missing information processing method based on multiple interpolation and Logistic regression [J]. Information Technology and Informatization, 2024, (08): 44-48.
[5] Miao Weijie, Wu Wenyuan. Privacy protection logic regression model based on homomorphic encryption training program [J/OL]. Computer engineering, 1-12 [2024-09-09]. https://doi.org/10.19678/j.issn.1000-3428.0069639.
[6] WANG Jinting, WANG Shouyang, Zhu Sheng. Risk analysis of queuing Game in Service system under mean-variance criterion [J/OL]. Science in China: Mathematics, 1-14 [2024-09-09]. http://kns.cnki.net/kcms/detail/11.5836.O1.20240807.1541.002.html.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.