Spatiotemporal Modeling and Resolution Enhancement for Aggregate Urban Mobility Analysis
実施中
瀬崎 薫
Urban mobility data are important for transportation planning, public-service allocation, and urban management. In many practical scenarios, mobility observations are available only at coarse spatial or temporal resolutions, or as incomplete samples, which makes it difficult to analyze fine-grained mobility patterns reliably. Real-world people-flow data provide an important benchmark for studying aggregate urban mobility dynamics and evaluating data-driven spatiotemporal modeling methods.This study aims to use the Real People Flow Data as a research benchmark for learning-based spatiotemporal modeling of aggregate mobility patterns. We will evaluate methods that estimate or enhance mobility distributions under reduced-resolution or sparse-observation settings and compare their performance across areas and time periods. The analysis will be conducted only at an aggregate and statistical level. We will not attempt to identify, track, or re-identify any individual.
変更のために新しい申請を保存します。 This will save a new application on the system for a modification.
申請中の研究者は表示されません。 / Pending researchers are not shown.
Han Zengyi / Jilin University
申請中のデータセットは表示されません。 / Pending datasets are not shown.
【別途書類手続き。通常より審査期間が長くなります】実人流データ（東京都、2025年5月）
年次報告の内容はメンバーのみ表示されます。