Census maps contain important socio-economic information regarding the population of a country. Computing these maps is critical given that policy makers often times make important decisions based upon such information. However, the compilation of census maps requires extensive resources and becomes highly expensive, especially for emerging economies with limited budgets. On the other hand, the ubiquitous presence of cell phones, both in developed and emerging economies, is generating large amounts of digital footprints. These footprints can reveal human behavioral traits related to specific socioeconomic characteristics. In this paper we propose a new tool, CenCell, to approximate census information from behavioral patterns collected through cell phone call records. The tool provides affordable census information by accurately classifying socioeconomic levels from cell phone call records with classification rates of up to 70%.
[1]
Bf Buxton,et al.
An introduction to support vector machines for data mining
,
2001
.
[2]
Nathan Eagle,et al.
Behavioral Inference across Cultures: Using Telephones as a Cultural Lens
,
2008,
IEEE Intelligent Systems.
[3]
Jesus Virseda,et al.
Towards large scale technology impact analyses: automatic residential localization from mobile phone-call data
,
2010,
ICTD 2010.
[4]
Víctor Soto,et al.
Prediction of socioeconomic levels using cell phone records
,
2011,
UMAP'11.
[5]
Vanessa Frías-Martínez,et al.
On the relationship between socio-economic factors and cell phone usage
,
2012,
ICTD.