Discovering urban spatial-temporal structure from human activity patterns

Urban geographers, planners, and economists have long been studying urban spatial structure to understand the development of cities. Statistical and data mining techniques, as proposed in this paper, go a long way in improving our knowledge about human activities extracted from travel surveys. As of today, most urban simulators have not yet incorporated the various types of individuals by their daily activities. In this work, we detect clusters of individuals by daily activity patterns, integrated with their usage of space and time, and show that daily routines can be highly predictable, with clear differences depending on the group, e.g. students vs. part time workers. This analysis presents the basis to capture collective activities at large scales and expand our perception of urban structure from the spatial dimension to spatial-temporal dimension. It will be helpful for planers to understand how individuals utilize time and interact with urban space in metropolitan areas and crucial for the design of sustainable cities in the future. Categories and Subject Descriptors H.2.8 [Database Applications]: data mining, spatial databases and GIS; I.5.3 [Clustering]: clustering algorithms and similarity measures