Multi-class identification of urban bus bunching rate based on XGBoost

Qian Liu; Mei Xiao; Xiuling Ming; Hongtao Huang

doi:10.1117/12.2658598

22 December 2022 Multi-class identification of urban bus bunching rate based on XGBoost

Qian Liu, Mei Xiao, Xiuling Ming, Hongtao Huang

Proceedings Volume 12460, International Conference on Smart Transportation and City Engineering (STCE 2022); 124603K (2022) https://doi.org/10.1117/12.2658598
Event: International Conference on Smart Transportation and City Engineering (STCE 2022), 2022, Chongqing, China

Abstract

It is of great practical significance to identify the state of bus bunching in advance so as to take reasonable measures. In order to improve the identification performance, a identification model based on Extreme Gradient Boosting (XGBoost) is proposed for multi classification of bus bunching rate. Firstly, using variance filtering and recursive feature elimination to screen the factors affecting the bus bunching rate. Secondly, SMOTE algorithm is used to deal with the data imbalance. Finally, XGBoost model is used to identify the multi classification of the bus bunching rate, and this paper compares the proposed model with other models. The research shows that the XGBoost model proposed in this paper has the best results in measurement indicators of identification performance, which verifies the applicability of the model to accurately identify the categories of bus bunching rate.

Citation Download Citation

Qian Liu, Mei Xiao, Xiuling Ming, and Hongtao Huang "Multi-class identification of urban bus bunching rate based on XGBoost", Proc. SPIE 12460, International Conference on Smart Transportation and City Engineering (STCE 2022), 124603K (22 December 2022); https://doi.org/10.1117/12.2658598

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available