Place ambiance has a huge influence over how we perceive places. Despite its importance, ambiance has been crucially overlooked by urban planners and scientists alike, not least because it is difficult to record and analyze at scale. We explored the possibility of using social media data to reliably map the ambiance of neighborhoods in the entire city of London. To this end, we collected geo-referenced picture tags from Flickr and matched those tags with the words in a newly created ambiance dictionary. In so doing, we made four main contributions: i) map the ambiance of London neighborhoods; ii) ascertain that such a mapping meets residents' expectations, which are derived from a survey we conducted; iii) show that computer vision techniques upon geo-referenced pictures are of predictive power for neighborhood ambiance; and iv) explain each prediction of a neighborhood's ambiance by identifying the picture that best reflects the meaning of that ambiance (e.g., artsy) in that neighborhood (e.g., South Kensington---the richest and most traditional neighborhood---and Shoreditch---among the most progressive and hipster neighborhoods in the city---are both 'artsy' but in very different ways). The combination of the predictive power of mapping ambiance from images and the ability to explain those predictions makes it possible to discover hidden gems across the city at an unprecedented scale.

The Spirit of the City: Using Social Media to Capture Neighborhood Ambiance

Schifanella, Rossano;
2018-01-01

Abstract

Place ambiance has a huge influence over how we perceive places. Despite its importance, ambiance has been crucially overlooked by urban planners and scientists alike, not least because it is difficult to record and analyze at scale. We explored the possibility of using social media data to reliably map the ambiance of neighborhoods in the entire city of London. To this end, we collected geo-referenced picture tags from Flickr and matched those tags with the words in a newly created ambiance dictionary. In so doing, we made four main contributions: i) map the ambiance of London neighborhoods; ii) ascertain that such a mapping meets residents' expectations, which are derived from a survey we conducted; iii) show that computer vision techniques upon geo-referenced pictures are of predictive power for neighborhood ambiance; and iv) explain each prediction of a neighborhood's ambiance by identifying the picture that best reflects the meaning of that ambiance (e.g., artsy) in that neighborhood (e.g., South Kensington---the richest and most traditional neighborhood---and Shoreditch---among the most progressive and hipster neighborhoods in the city---are both 'artsy' but in very different ways). The combination of the predictive power of mapping ambiance from images and the ability to explain those predictions makes it possible to discover hidden gems across the city at an unprecedented scale.
2018
2
CSCW
144:1
144:18
https://dl.acm.org/citation.cfm?id=3274413
ambiance, computer vision, flickr, london
Redi, Miriam; Aiello, Luca Maria; Schifanella, Rossano; Quercia, Daniele
File in questo prodotto:
File Dimensione Formato  
redi18spirit.pdf

Accesso riservato

Tipo di file: PDF EDITORIALE
Dimensione 2.09 MB
Formato Adobe PDF
2.09 MB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/2318/1726558
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 21
  • ???jsp.display-item.citation.isi??? ND
social impact