Education, Science, Technology, Innovation and Life
Open Access
Sign In

Research on Used Car Transaction Cycle Based on Soft Voting

Download as PDF

DOI: 10.23977/infse.2022.030103 | Downloads: 11 | Views: 898

Author(s)

Siyu Zhao 1, Kangan Qian 2, Chengge Wen 3

Affiliation(s)

1 School of Resources and Safety Engineering, Central South University, Hunan, Changsha, 410083, China
2 College of Mechanical and Electrical Engineering, Central South University, Hunan, Changsha, 410083, China
3 Bussiness School of Central South University, Hunan Changsha, 410083, China

Corresponding Author

Siyu Zhao

ABSTRACT

In this paper, the key factors affecting the vehicle transaction cycle are mined, and the soft voting integrated learning algorithm model is established to solve it. Logistic regression, AdaBoost, gbdt, SVM and random forest are considered as base classifiers. Firstly, 24 candidate indexes are determined as candidate key factors affecting vehicle transaction cycle Then, data fitting and parameter adjustment are carried out for the five base classifiers, and the better parameters are selected to obtain the performance of the five base classifiers on the data set. The weight of each base classifier is judged by the comprehensive accuracy of the classifier on the overall data. Finally, through the descriptive statistical analysis of the candidate index data through the soft voting model, the indicators with an impact index of more than 90% on the vehicle transaction cycle are extracted as the result, which are the price adjustment time: adjusted price, exhibition time, model id, city id of the vehicle, transfer times, country, fuel type, new car price. Based on this, the effective means to speed up sales are put forward.

KEYWORDS

Used car, Soft Voting, transaction cycle

CITE THIS PAPER

Siyu Zhao, Kangan Qian, Chengge Wen, Research on Used Car Transaction Cycle Based on Soft Voting. Information Systems and Economics (2022) Vol. 3: 10-15. DOI: http://dx.doi.org/10.23977/infse.2022.030103.

REFERENCES

[1] Dongfang fortune com.2021 Analysis on the development prospect of China's second-hand car testing market in [EB/OL]. baijiahao. baidu. com/s? id=1707777601753410628& wfr=spider& for=pc, 2021­8­11.
[2] Meng ye, Yu Zhongqing, Zhou Qiang. Stock index prediction method based on Ensemble Learning[J]. modern electronic technology, 2019,42 (19): 115 ­ 118. DOI: 10.16652/j.issn. one thousand and four ­ 373x. 2019.19.027.
[3] Wang Jinzhu, Wang Xiang. Learn Python data analysis and mining from scratch[M]. Beijing: Tsinghua University Press, 2018.

Downloads: 6147
Visits: 130655

All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.