Sensing Human Society

Shoval, N. (2007). Sensing human society. Environment and Planning B: Planning and Design, 34:191–195.

In this paper Noam Shoval proposes referring to the possibility of the aggregative use of cell phone spatial data as human sensing. He makes the distinction with remote sensing that applies more to the domain of physical geography arguing that human sensing enables us to sense people directly, in contrast to remote sensing, where the information obtained regarding human activity is indirect. Even if new sources of physical geographic data is related to aspects of human spatial activity, none of them can be thought of as materially augmenting more traditional social scientific data that describe the social characteristics of individuals and groups. Now the emergence of human sensing coupled with the considerable progress in the field of GIS, currently places us on the verge of a veritable revolution in human time-space activity research.
The author introduces the two methods of analysis of location data of cell phone that I have explored in the course of my thesis. First from the activity of devices that are associated with each antenna of the network during a certain period of time. Second, the detection of the location and migration of devices over a given period of time on the different antennas of the networks. Both approaches raise privacy issues, even though Noam mentions that “the data derive from statistics regarding activity in antennas and not regarding the locations of the phones themselvesötherefore there should not be any privacy considerations in this case”, comparing the approach to study census data for which “detailed census records at the individual household level are not usually available to researchers. Data are available only in aggregated forms at a level of detail intended to prevent the ready association of demographic information with individual households.”

Other challenges confirms the limitations that have emerged from my thesis works: a) aggregated cellphone data are particularly difficult to deal with high-resolution
interurban analysis; being more appropriate for studies of relatively low geo-graphical resolution, such as at the metropolitan level b) the representativeness of the cellphone data than necessitate calibration methods (although such issues must be addressed in any other research method as well)

Interestingly, Noam suggests the observation cellular network infrastructure already reveals general idea about the main nodes of activity in cities, such as concentrations of business activities, shopping, and leisure, such as here for Manhattan:

Image courtesy of Noam Shoval (original), created by Adi Ben-Nun from the Hebrew University GIS Centre, of a density map of around 1300 cellular antennas in Manhattan.
Relation to my thesis: “Time will tell whether the contribution made to human geography and the social sciences with this methodology will be on a par with the revolution generated by remote sensing in the field of physical geography and related geosciences.“