We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Short-term BRT Passenger Flow Prediction with A Deep Learning Method.
- Authors
Haitao XU; Jiaxue PENG; Ning ZHENG; Ying GAO
- Abstract
Predicting passenger flow accurately can make full use of the existing traffic facilities to prevent emergencies from happening which is caused by heavy passenger flow. However, most of the existing methods to predict short-term passenger flow are shallow architecture, they cannot mine the relationship between short-term passenger and other external factors deeply enough. Therefore, we consider using a deep learning method to predict Bus Rapid Transit (BRT) passenger flow, which is different from previous methods to predict BRT passenger flow. We use a Prediction Stacked Auto Encoder (PSAE) model to mine BRT passenger flow features deeply. The top of the whole model establishes a prediction model for BRT passenger flow prediction from the extracted feature set. It uses a greedy layer-wise algorithm to train model parameters. Experimental results show that this method has good performance for predicting short-term BRT passenger flow.
- Subjects
BUS rapid transit; PASSENGERS
- Publication
International Journal of Simulation: Systems, Science & Technology, 2016, Vol 17, Issue 40, p1
- ISSN
1473-8031
- Publication type
Article
- DOI
10.5013/IJSSST.a.17.40.23