Interpretable train-level entry flow prediction using GMM-based features and ensemble learning

doi:10.1016/j.asoc.2025.114357

Interpretable train-level entry flow prediction using GMM-based features and ensemble learning

2025-12-03

0

PRE

AI

王增强 (Zengqiang Wang) *

F

Francisco Chiclana

DOI：10.1016/j.asoc.2025.114357

Original

Original request for help

Save

Abstract

En 中文

• Introduced a two-stage framework for train-level passenger flow prediction. • Incorporated passenger heterogeneity for enhanced prediction accuracy. • Achieved 70 % accuracy improvement using a two-stage prediction strategy. • Utilized GMM to capture nuanced train-level passenger flow distributions. • SHAP analysis revealed key factors influencing passenger flow patterns.

Interpretable train-level entry flow prediction using GMM-based features and ensemble learning

Abstract

Journal

Organization