Research map / Graph learning / Knowledge and temporal graphs
Online Metro Origin-Destination Prediction via Heterogeneous Information Aggregation
Lingbo Liu, Yuying Zhu, Guanbin Li, Ziyi Wu, Lei Bai, Liang Lin
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add unfinished order vectors without destination estimation (ablation) | IOD-Net · SHMOD OD 60 min | MAPE (lower better): 42.43% -> 41.06% | Improved |
| add unfinished order vectors without destination estimation (ablation) | IOD-Net · HZMOD OD 60 min | MAPE (lower better): 35.11% -> 32.60% | Improved |
| estimate destinations of unfinished orders with long and short-term distributions (ablation) | IOD+U-Net · HZMOD OD 60 min | MAPE (lower better): 32.60% -> 30.05% | Improved |
| estimate destinations with long-term distribution only (ablation) | IOD+U-Net · SHMOD OD 60 min | MAPE (lower better): 41.06% -> 40.43% | Improved |
| add OD-to-DO input (IOD+U(short+long)) on top of DO history (ablation) | DO to DO · SHMOD DO 60 min | MAPE (lower better): 42.05% -> 39.18% | Improved |
| use unfinished orders to forecast DO instead of DO history alone (ablation) | DO to DO · SHMOD DO 60 min | MAPE (lower better): 42.05% -> 40.54% | Improved |
| add DO-to-OD correlation to IOD+U(short+long) input (ablation) | OD prediction · SHMOD OD 60 min | MAPE (lower better): 39.18% -> 38.36% | Improved |
| replace no interaction with cross-station DIT interaction (ablation) | HIAM · SHMOD DO 60 min | MAPE (lower better): 42.05% -> 39.18% | Improved |
| replace single-station interaction with DIT (ablation) | HIAM · SHMOD OD 60 min | MAPE (lower better): 39.24% -> 38.36% | Improved |
| add single-station interaction (ablation) | HIAM · SHMOD OD 60 min | MAPE (lower better): 39.18% -> 39.24% | No clear effect |
| increase input sequence length from 1 to 4 (ablation) | HIAM · SHMOD OD 60 min | MAPE (lower better): 39.46% -> 38.36% | Improved |
| increase input sequence length from 4 to 5 (ablation) | HIAM · SHMOD OD 60 min | MAPE (lower better): 38.36% -> 38.39% | No clear effect |
Closest papers by meaning
- SeoulMMOD: A Large-Scale Multimodal Origin-Destination Flow Benchmark (NeurIPS 2026 · similarity 0.86)
- Generating Origin-Destination Matrices in Neural Spatial Interaction Models (NeurIPS 2024 · similarity 0.84)
- Responsive Dynamic Graph Disentanglement for Metro Flow Forecasting (AAAI 2025 · similarity 0.83)
- BO4Mob: Bayesian Optimization Benchmarks for High-Dimensional Urban Mobility Problem (NeurIPS 2025 · similarity 0.83)
- Disentanglement-Guided Spatial-Temporal Graph Neural Network for Metro Flow Forecasting (Student Abstract) (AAAI 2024 · similarity 0.82)
- Satellites Reveal Mobility: A Commuting Origin-destination Flow Generator for Global Cities (NeurIPS 2025 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
