Human mobility data acquired by smartphone applications or cell phone networks enable us to observe people's movements and activities quantitatively and dynamically. Using these technologies, I have developed a method to estimate where and how much people walk based on movement trajectories. This method can be used to assess the impact of unexpected events, such as natural disasters or infectious disease outbreaks, on people's daily activity.
As with the people's activities, the urban environment can be observed from a variety of spatial big data. By using deep learning technique, it is possible to extract segmentations of images for streetscape to quantitatively understand the state of the landscape at specific locations. In my previous work, I have built a model to automatically assess whether the streetscape at a given location is walking-friendly for older people or not.
Nagata, S. et al. (2020). Objective scoring of streetscape walkability related to leisure walking: Statistical modeling approach with semantic segmentation of Google Street View images. Health & Place, 66, 102428, doi:10.1016/j.healthplace.2020.102428
Nagata, S. et al. (2021). Mobility Change and COVID-19 in Japan: Mobile Data Analysis of Locations of Infection. Journal of Epidemiology, 31(6), 387–391, doi:10.2188/jea.JE20200625
Nagata, S. et al. (2021). Relationships among changes in walking and sedentary behaviors, individual attributes, changes in work situation, and anxiety during the COVID-19 pandemic in Japan. Preventive Medicine Reports, 24, 101640, doi:10.1016/j.pmedr.2021.101640
Nagata, S. et al. (2022). Development of a method for walking step observation based on large-scale GPS data. International Journal of Health Geographics, 21(1), 10, doi:10.1186/s12942-022-00312-5