Analysis of the Influencing Factors Affecting the Satisfaction of New Energy Vehicles Based on CatBoost Model

Weiguang Jin; Shilan Hu; Shouxi Wu

doi:10.62051/ijgem.v7n1.04

Authors

Weiguang Jin
Shilan Hu
Shouxi Wu

DOI:

https://doi.org/10.62051/ijgem.v7n1.04

Keywords:

CatBoost, SPSS, Box plot, Min-Max, Standardization

Abstract

In order to solve the problem that automobile enterprises want to be able to keep track of the satisfaction of a large number of consumers with new energy vehicle products at all times, this paper proposes a scheme based on the CatBoost machine learning model to analyze the influencing factors of the satisfaction of new energy vehicle products. This scheme first uses market research methods and SPSS software to obtain the relevant original data; Next, the box plot is used to find outliers, the R language function is used to find missing values, and the missing values are filled with mean and mode respectively according to the characteristics of the dimension data, and then the Min-Max method is used to standardize the data. Finally, the principle of the CatBoost model is introduced; The results show that range, safety and economy are the indicators that consumers are more concerned about, and the results meet the actual requirements. Based on the research conclusions of this paper, it can provide reference and basis for improving product satisfaction and precision marketing of enterprises.

Downloads

Download data is not yet available.

References

[1] H. Huiying, Y. Jing, Z. Fan and Z. Yishu, "Brand evaluation system for vehicles based on multi-source heterogeneous data and CatBoost model," 2024 IEEE 2nd International Conference on Electrical, Automation and Computer Engineering (ICEACE), Changchun, China, 2024, Pp. 670-676, doi: 10.1109 / ICEACE63551.2024.10898605.

[2] Li Yuhan, Shi Zhen. Evaluation of riding satisfaction of electric vehicles based on Principal component analysis [J]. Automobile Practical Technology, 2020(11):8-10.

[3] Dongfang Yuxiao. The prospect of new energy vehicles is broad, but consumer satisfaction needs to be improved [J]. China World, 2019(07):58-59.

[4] Du Jiarui, Gao Lequan, Liu Yan, Li Bin, Yang Zhongmin. Research on Influencing Factors of Purchase Intention of New Energy Vehicle Consumers in Beijing-Tianjin-Hebei [J]. Cheng, 2019, 38(19):220-223.

[5] Yu Jicheng, Zhou Feng, Wang Jiangchu, et al. Prediction of transmission line loss rate based on multi-dimensional features and GBDT model [J]. Computer Applications and Software, 2022, 39 (06): 82-86+126.

[6] Zhang Fan, Guo Yaxin, Yang Jing, et al. Research on prediction of electric quantity based on GBDT+ feature engineering method [J]. Electron Quality, 2020, (01): 1-4.

[7] Fan Zhang et al 2021 IOP Conf. Ser.: Earth Environ. Sci. 7669 042023DOI 10.1088/1755-1315/769/4/042023.

[8] Zhang Yixiao, Zhao Zhongguo, Zheng Jianghua. CatBoost: A new approach for estimating daily reference crop evapotranspiration in arid and semi-arid regions of Northern China [J]. Journal of Hydrology, 2020(prepublish).

[9] Ding Qi. Research on Employee Departure Prediction Based on Catboost algorithm [D]. Shanghai Normal University, 2020.

[10] Dang Cunlu, Wu Wencheng, Li Chaofeng, Li Yongqiang. Research on short-term power load forecasting based on CatBoost algorithm [J]. Journal of Electrical Engineering, 2020, 15 (01):76-82.

[11] Jiang Qigang, Yang Xiuyan, Yang Changbao, Zhao Zhenhe. Object-oriented land use classification based on CatBoost algorithm [J]. Journal of Jilin University (Information Science Edition), 2020, 38(02):185-191.