Resembling population density distribution with massive mobile phone data

Abstract

As the mobile phone data (CDR data) has gained an increasing interest in research, such as social science, transportation, urban informatics, and big data, this study aims at examining the representativeness of the CDR data in terms of resemblance of the actual population density distribution from three perspectives; operator’s market share, urban-rural user population ratio, and user gender ratio. The results reveal that the representativeness of the data does not scale at the same rate with the operator’s market share, the urban-rural user population ratio of 80:20 can best represent the population density distribution, and an equal mixture of male and female user population can best resemble the population density distribution. This study is the first investigation into the representativeness of the CDR data. The findings provide useful information, which can serve an insightful guideline when dealing with the CDR data.