Improving land use inference by factorizing mobile phone call activity matrix

Abstract

Land use is shaped by human activities. Traditional methods of measuring land uses (e.g. surveys and remote sensing techniques) often have diculties in capturing human activities. The near-ubiquitous coverage of mobile phones opens up a new way to investigate land use through human activities. We propose to analyze land use by characterizing human activity patterns based on the aggregated call volume, and apply non-negative matrix factorization to identify fundamental behavioral classes. Using tower-based call data from Dakar, Senegal, we discover two fundamental land use patterns: commercial/business/industrial and residential. The land use of the reception area of each cell tower can be inferred based on the weights obtained for each basis vector.We evaluate our results using the Points-of-Interest datasets ob- tained from OpenStreetMap and Facebook's Graph API. Our results demonstrate the promise of our method for inferring mixed land use and tracking land use changes in real time.