A novel traffic system called bike sharing system (BSS) has appeared in daily life without much fanfare. It is surprising that people tend to use BSS to make their life more convenient. However, it also brings traffic problems such as mass intersection situations, and illegal riding. As a result, how to predict bike usage of BSS in the short-term under various weather conditions becomes a significant traffic problem. In this paper, future bike usage can be predicted based on historic and weather data collected in Seattle. Models are established by machine learning algorithms; and a suitable feature selection method is proposed. Performance indexes will be used to evaluate each model.
Weather Impact on a Bike Sharing System Based on Machine Learning
18th COTA International Conference of Transportation Professionals ; 2018 ; Beijing, China
CICTP 2018 ; 329-338
02.07.2018
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Exploring the weather impact on bike sharing usage through a clustering analysis
ArXiv | 2020
|