Short-Term Metro Daily Passenger Flow Prediction Using Machine Learning

Leyang Liu

2023

Abstract

The prediction of daily passenger flow in the metro would be meaningful to the construction and operation of urban rail transit, which is common in megacities of China. The study takes the daily passenger flow of the Beijing metro as an example and tries to make a short-term prediction of it based on its historical data. Since the data volume is relatively small, resulting in an overfitting problem when applying mainstream time series models like Recurrent Neural Network (RNN) and Long Short Term Memory (LSTM), four selected machine learning models are applied to this topic. Their prediction performance is compared by not only the common indicator like Mean Squared Error (MSE) and their comprehensive performance. The result shows that the machine learning models considering both seasonality and holiday factors perform best and have the strongest interpretability. For future research, it’s possible that the combination of multiple machine learning models would achieve better results or with stronger interpretability in this topic.

Paper Citation

in Harvard Style

Liu L. (2023). Short-Term Metro Daily Passenger Flow Prediction Using Machine Learning. In Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML; ISBN 978-989-758-705-4, SciTePress, pages 349-355. DOI: 10.5220/0012809800003885

in Bibtex Style

@conference{daml23,
author={Leyang Liu},
title={Short-Term Metro Daily Passenger Flow Prediction Using Machine Learning},
booktitle={Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML},
year={2023},
pages={349-355},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012809800003885},
isbn={978-989-758-705-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML
TI - Short-Term Metro Daily Passenger Flow Prediction Using Machine Learning
SN - 978-989-758-705-4
AU - Liu L.
PY - 2023
SP - 349
EP - 355
DO - 10.5220/0012809800003885
PB - SciTePress