Space-time clustering for identifying population patterns from smart-phone data

Lucia Paci, Francesco Finazzi

Risultato della ricerca: Contributo in libroContributo a convegno

Abstract

In this work we aim at studying spatio-temporal patterns of the population movement across a large city. We exploit the information on people position collected by the smartphone application of the Earthquake Network project and we adopt a dynamic model-based clustering approach to identify the patterns. The approach is applied to smartphone data collected in Santiago (Chile) over the period February-April 2016. Some preliminary results are presented and discussed.
Titolo tradotto del contributo[Autom. eng. transl.] Space-time clustering for identifying population patterns from smart-phone data
Lingua originaleItalian
Titolo della pubblicazione ospiteSIS 2017. Statistics and Data Science: New Challenges, New Generations
Pagine1-6
Numero di pagine6
Stato di pubblicazionePubblicato - 2017
EventoSIS 2017. Statistics and Data Science: New Challenges, New Generations - Firenze
Durata: 28 giu 201730 giu 2017

Convegno

ConvegnoSIS 2017. Statistics and Data Science: New Challenges, New Generations
CittàFirenze
Periodo28/6/1730/6/17

Keywords

  • Finite mixture models
  • Markov chain Monte Carlo
  • crowd-sourcing data
  • spatio-temporal modeling
  • state-space

Fingerprint Entra nei temi di ricerca di 'Space-time clustering for identifying population patterns from smart-phone data'. Insieme formano una fingerprint unica.

Cita questo