Crowdsourcing Pleasantness and Safety Perceptions: An Analysis through Multiple Rankings and Socio-Demographic Groups

ABSTRACT

Our perception of public spaces guides our behavior on them. Understanding which factors shape this perception informs both urban planners, who aim to improve city life, as well as computational models that help us navigate urban spaces. Crowdsourcing games have been employed to evaluate citizens’ perceptions of urban scenes at scale. This work contributes to crowdsourcing studies by evaluating different strategies to aggregate perceptions and rank scenes, as well as investigating the relation of captured perceptions with features of the urban form and the socio-demographic profile of participants.

KEYWORDS:

• urban informatics
• urban perception
• crowdsourcing

Declaration of Interest

The authors declare no conflict of interest.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 https://contribua.org/project/comoecampinav2/

2 https://contribua.org

3 http://pybossa.com/

4 https://www.google.com.br/intl/en/streetview/

5 http://www.newsweek.com/new-brand-tech-cities-150557

6 https://github.com/anishathalye/gavel

7 https://github.com/fivethirtyeight/nfl-elo-game

8 https://webfiles.uci.edu/kday/public/index.html

9 https://gehlinstitute.org/news/launching-open-public-life-data-protocol-version-1-0/

10 http://www.crowdflower.com/

Additional information

Notes on contributors

David Candeia

David Candeia is a professor at Instituto Federal de Educação, Ciência e Tecnologia da Paraíba, Brazil. His research interests include urban informatics, urban computing, cloud computing and peer-to-peer systems.

Flavio Figueiredo

Flavio Figueiredo is a professor at Universidade Federal de Minas Gerais (UFMG). He has been a visiting scholar at Carnegie Mellon University as well as the at the University of British Columbia. He also worked for a year at IBM's Research Lab in Rio de Janeiro. Currently, he performs research developing and applying data science and machine learning algorithms for a wide range of contexts (from social media, Internet traces, and cultural production).

Nazareno Andrade

Nazareno Andrade is a professor at Universidade Federal de Campina Grande (UFCG). During his PhD he was a visiting scholar at University of British Columbia and Hewlett Packard Labs Bristol, as well as Delft University of Technology after completing his PhD. His research focuses on applied data science—focusing on urban issues, music computing and other contexts—data visualization and machine learning.