Sentiment Analysis-Based Subway Passenger Flow Prediction and Decision Support

Yue Yu

2025

Abstract

With the acceleration of urbanization, subways have become vital to public transportation. Accurate passenger flow prediction is key to optimizing operations and improving service. Traditional methods of prediction mainly use historical data, often ignoring emotional factors. Using machine learning algorithms and social media sentiment analysis, this study investigates how public emotion affects subway passenger flow. First, web scraping techniques are used to collect data from Weibo, and the large language model is used for sentiment analysis. Then, a random forest model is constructed using both sentiment data and historical subway passenger flow data for prediction. Experimental results indicate that model incorporating sentiment data is more accurate in predictions than traditional methods, particularly during emergencies or special time periods. This study provides a new perspective for subway management, which can be applied to optimize scheduling strategies.

Paper Citation

in Harvard Style

Yu Y. (2025). Sentiment Analysis-Based Subway Passenger Flow Prediction and Decision Support. In Proceedings of the 2nd International Conference on Public Relations and Media Communication - Volume 1: PRMC; ISBN 978-989-758-778-8, SciTePress, pages 351-360. DOI: 10.5220/0013991700004916

in Bibtex Style

@conference{prmc25,
author={Yue Yu},
title={Sentiment Analysis-Based Subway Passenger Flow Prediction and Decision Support},
booktitle={Proceedings of the 2nd International Conference on Public Relations and Media Communication - Volume 1: PRMC},
year={2025},
pages={351-360},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013991700004916},
isbn={978-989-758-778-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 2nd International Conference on Public Relations and Media Communication - Volume 1: PRMC
TI - Sentiment Analysis-Based Subway Passenger Flow Prediction and Decision Support
SN - 978-989-758-778-8
AU - Yu Y.
PY - 2025
SP - 351
EP - 360
DO - 10.5220/0013991700004916
PB - SciTePress