SeoulMMOD: A Large-Scale Multimodal Origin-Destination Flow Benchmark

Key points

• We introduce SeoulMMOD, a large-scale public benchmark for citywide multimodal mobility in Seoul.1
• SeoulMMOD provides three years of hourly OD flows estimating total mobility for 6 urban travel modes across 25 districts and 426 sub-districts, together with travel-time, travel-distance, calendar, rainfall, and point-of-interest (POI) signals.1
• Using SeoulMMOD, we benchmark 14 forecasting baselines under a common protocol and analyse the effects of spatial scaling, joint versus mode-separate training, and cross-year generalization.1
• Results show that several spatio-temporal models that work at district level do not scale to sub-district OD forecasting, and that naive joint multi-mode training does not consistently improve accuracy.1

Sources (1)

• [1]SeoulMMOD: A Large-Scale Multimodal Origin-Destination Flow Benchmark
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 9, 03:27 PM

  Origin-destination (OD) flow forecasting supports urban mobility decisions by predicting the volume and patterns of human movement across a city.

  We introduce SeoulMMOD, a large-scale public benchmark for citywide multimodal mobility in Seoul.

Extractive summary: sentences quoted from the sources.