Inferring Urban Population Distribution from GSM Data: An Experimental Case Study

摘要

A huge amount of location and tracking data is gathered by tracking and location technologies, such as Global Positioning system (GPS) and Global System for Mobile communication (GSM) devices leading to the collection of large spatiotemporal datasets and to the opportunity of discovering usable knowledge about movement behavior. Movement behavior can be extremely useful in many ways when applied, for example, in the domain of planning metropolitan areas, traffic management, mobile marketing, tourism, etc. Currently, population statistics are derived from census data and household surveys. However, they are difficult to manage and costly to implement. In this paper, we move towards this direction and propose a technique for inferring urban population distribution from GSM data. A case study is used from to build a prototype for testing and evaluating the proposed technique. Results showed that we were able to infer population density with high accuracy based on analyzing GSM data.