A multicriteria optimization framework for urban social media analytics

Monday 29 November 2021
14:00 (GMT)
Online (via Microsoft Teams)

Professor João Porto de Albuquerque, Deputy Director of UBDC, will present at this International Journal of Geographical Information Science (IJGIS) webinar.

Professor João Porto de Albuquerque and Professor Sidgley Camargo de Andrade, Federal University of Technology - Paraná, will discuss their paper which takes multicriteria optimization to spatial analysis of social media data and determines the spatial area unit that maximizes spatial autocorrelation. The paper is freely accessible.

In this talk they will discuss the challenge of determining the most suitable areal unit for the spatial analysis of georeferenced social media messages, as a recent instantiation of the classic Modifiable Areal Unit Problem (MAUP). They will present their approach to support the definition of spatial granularity based on multi-objective optimization applied to spatial autocorrelation indexes and discuss its application in the context of social media responses to rainfall in the city of Sao Paulo, Brazil. They will conclude with a broader discussion of other applications and further avenues of investigation, including their recent follow-on IJGIS paper on the effect of intra-urban mobility flows on the spatial heterogeneity of social media activity.