Identifying latent activity behaviors and lifestyles using mobility data to describe urban dynamics

Abstract

Urbanization and its problems require an in-depth and comprehensive understanding of urban dynamics, especially the complex and diversified lifestyles in modern cities. Digitally acquired data can accurately capture complex human activity, but it lacks the interpretability of demographic data. In this paper, we study a privacy-enhanced dataset of the mobility visitation patterns of 1.2 million people to 1.1 million places in 11 metro areas in the U.S. to detect the latent mobility behaviors and lifestyles in the largest American cities. Despite the considerable complexity of mobility visitations, we found that lifestyles can be automatically decomposed into only 12 latent interpretable activity behaviors on how people combine shopping, eating, working, or using their free time. Rather than describing individuals with a single lifestyle, we find that city dwellers’ behavior is a mixture of those behaviors. Those detected latent activity behaviors are equally present across cities and cannot be fully explained by main demographic features. Finally, we find those latent behaviors are associated with dynamics like experienced income segregation, transportation, or healthy behaviors in cities, even after controlling for demographic features. Our results signal the importance of complementing traditional census data with activity behaviors to understand urban dynamics.

Home-to-school pedestrian mobility GPS data from a citizen science experiment in the Barcelona area

“…Then, based on the disclosure risk 1/ k willing to assume, we can finally decide the time period t in seconds we strip away from indidvidual GPS records at the beginning of the home-to-school trajectories (or at the end for school-to-home trajectories, but always when participants have non-zero velocity). If we approximately take v = 1.5 m/s for all participants (see Data Records Section to check that this is a reasonable choice) and use t = 50 s, we can obtain an average disclosure risk 〈1/ k 〉 = 2.72 × 10 −3 , which is a comparable order of magnitude taken by other publications using GPS data 6 – 8 . Table 3 shows the details for each of the districts and cities.…”

“…Apart from the velocity we show in Methods, it is possible to further characterize pedestrian mobility with other statistical metrics like reorientation angle or tortuosity 33 . Finally, it can also be of interest to correlate some of the statistical metrics with contextual information such as amount of green or width of the side-walks along similar lines to recent publications 6 – 8 . The model parameters could be estimated from the empirical data and used to compare mobility in the schools’ surroundings in terms of urban structure, more walkable routes (pedestrian streets and green areas) or climate conditions 35 .…”

“…Much research has been done to understand multi-modal transportation in our cities, especially in the last ten years, due to the availability of digital traces 4 . However, although digital traces work well to investigate large distance movements or origin-destination flows (typical public transportation or vehicles) 5 , they are less precise in investigating the local micro-behaviours around people’s homes or in pedestrian areas 6 – 8 . There is an increasing need to get more precise data to better learn about the micro-motives of what makes an area more suitable for walking, how different spaces are used to walk, or what activities are more impacted by walking accessibility.…”

“…Mobility data must avoid participants identification by for instance inferring home address (see for instance refs. 6 – 8 ) and geomasking with for instance spatial k -anonimity technique is required 17 , 18 . In citizen science projects, several aspects are also raised and some measures such as those taken CitSci.org platform could be required as well 19 .…”

How people are exposed to neighborhoods racially different from their own

“…No studies have either proposed a BCZ measure or examined what POI contribute to it. However, studies focused on income rather than race have already shown that establishments drive people to travel to distal parts of the city ( 25 – 28 ), that consumer purchasing close to the residence is associated with exposure to similar income groups ( 15 ; see also refs. 16 and 28 ), and that people tend to encounter other income groups when attending parks, libraries, and some restaurant chains ( 17 , 21 ; see also ref.…”