| Date | Thursday 9, April, 2026 12:40-13:40 |
|---|---|
| Venue | Room 205, Annex (*) 2nd floor (*) No. 5 building in the campus map https://www.hit-u.ac.jp/eng/about/direction/campusmap/kunitachi.html |
| Speaker | Peiran Li (Assistant Professor, HIAS) |
| Title | "Using Mobile Phone Data for Spatial Information Science: Generation, Inference, and Applications" |
| Abstract | Mobile phone location data have become an important data source for spatial information science, offering new opportunities to observe human activities, mobility patterns, and urban dynamics at unprecedented spatial and temporal scales. At the same time, their effective use raises several methodological challenges, including privacy constraints, limited data accessibility, and the difficulty of extracting socially meaningful information from raw trajectories. This talk presents three complementary lines of research addressing these challenges. First, I introduce recent work on pseudo trajectory generation, focusing on AI-driven generative model for GPS trajectory generation that aims to improve scalability, transportation-mode diversity, and generation efficiency under privacy-aware settings. Second, I discuss demographic inference from mobile phone trajectories, including a Bayesian approach for estimating age and gender patterns from anonymized mobility data and census information, with the goal of tracking demographic dynamics in built environments. Third, I highlight how mobile phone data can support urban applications by revealing behavioral patterns and social heterogeneity in cities. Taken together, these studies illustrate how mobile phone data can contribute not only to movement observation, but also to data generation, semantic inference, and evidence building for urban and spatial research. | Language | English |
| Note | Registration:https://forms.cloud.microsoft/r/yAA70Wa6G1 (Deadline: 3 PM, 8 April) *Bring your own lunch. Coffee and snacks will be served. |