Multi-step traffic flow prediction using stacking ensemble learning model

Shi Yin; Hui Liu; Yanfei Li; Jing Tan; Jiakang Wang

doi:10.1117/12.2620226

22 December 2021 Multi-step traffic flow prediction using stacking ensemble learning model

Shi Yin, Hui Liu, Yanfei Li, Jing Tan, Jiakang Wang

Proceedings Volume 12058, Fifth International Conference on Traffic Engineering and Transportation System (ICTETS 2021); 120582M (2021) https://doi.org/10.1117/12.2620226
Event: 5th International Conference on Traffic Engineering and Transportation System (ICTETS 2021), 2021, Chongqing, China

Abstract

The prediction of traffic flow can supply a powerful reference for urban traffic planning. A multi-step traffic flow prediction model using stacking ensemble learning algorithm is constructed in this paper, which ensembles three Elman Neural Networks (ENN) as base learn and a Support Vector Machine (SVM) as meta-learner. Experimental results prove that the good multi-step prediction effect can be achieved by the proposed stacking ensemble model, and compared with Autoregressive Integrated Moving Average (ARIMA), ENN, SVM and Echo State Networks (ESN), the proposed model can achieve the highest prediction accuracy regardless of 1, 2 or 3 step forecasting.

Citation Download Citation

Shi Yin, Hui Liu, Yanfei Li, Jing Tan, and Jiakang Wang "Multi-step traffic flow prediction using stacking ensemble learning model", Proc. SPIE 12058, Fifth International Conference on Traffic Engineering and Transportation System (ICTETS 2021), 120582M (22 December 2021); https://doi.org/10.1117/12.2620226

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