Airline Passenger Satisfaction Prediction and Key Influential Factors Identification Based on Logistic Regression and Random Forest

Jiayi Xue

2025

Abstract

The demand for tourism has grown rapidly since the post-pandemic reopening, and the tourism industry has ushered in a new wave of recovery. How airlines can provide a flight experience that satisfies travellers has once again become a matter of importance. This study aims to model airline passenger satisfaction and screen some of the factors that have the greatest impact on the level of satisfaction. First, dimensionality reduction of the dataset was realized through the principal component analysis method. The study then applied logistic regression and random forest algorithms and compared both results, using confusion matrices and various model metrics. It was found that the random forest performed better than the logistic regression algorithm, with an accuracy of 92% vs. 85%. This suggests that the Random Forest model is more suitable for this dataset. Then random forest model was applied to rank the importance of features. It turns out that the priority of digital experience services ranks high in the list, which also gives some indication of the direction of airline services improvement. Future research could introduce deep learning models to optimize the performance by fusing real-time data from multiple sources, while incorporating interpretable technologies to drive aviation services towards precision and personalization.

Paper Citation

in Harvard Style

Xue J. (2025). Airline Passenger Satisfaction Prediction and Key Influential Factors Identification Based on Logistic Regression and Random Forest. In Proceedings of the 2nd International Conference on Innovations in Applied Mathematics, Physics, and Astronomy - Volume 1: IAMPA; ISBN 978-989-758-774-0, SciTePress, pages 610-614. DOI: 10.5220/0013834200004708

in Bibtex Style

@conference{iampa25,
author={Jiayi Xue},
title={Airline Passenger Satisfaction Prediction and Key Influential Factors Identification Based on Logistic Regression and Random Forest},
booktitle={Proceedings of the 2nd International Conference on Innovations in Applied Mathematics, Physics, and Astronomy - Volume 1: IAMPA},
year={2025},
pages={610-614},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013834200004708},
isbn={978-989-758-774-0},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 2nd International Conference on Innovations in Applied Mathematics, Physics, and Astronomy - Volume 1: IAMPA
TI - Airline Passenger Satisfaction Prediction and Key Influential Factors Identification Based on Logistic Regression and Random Forest
SN - 978-989-758-774-0
AU - Xue J.
PY - 2025
SP - 610
EP - 614
DO - 10.5220/0013834200004708
PB - SciTePress