Accueil > Géo-visualisations

Representing uncertain futures

par Christine Voiron - publié le , mis à jour le

Representing Uncertain Futures : Social Polarization in the Metropolitan Area of Marseille

 

Giovanni FUSCO, Cristina Minh Thu CAO - UMR7300 ESPACE, CNRS / Univérsité Nice Sophia Antipolis

Analyses and Data-Viz realized within the PEPS Géo-Incertitude project (CNRS grant 2014-15).

The metropolitan area of Aix-Marseille in southern France has experienced ongoing social polarization since the 1980s. The geography of unemployment, on the one hand, and the concentration of high-skilled professionals, on the other, contribute considerably to the structuring of a contrasted metropolitan social morphology (Centi 1993, 1996, Fusco and Scarella 2011). Future continuation of metropolitan-wide logics, reinforcing social polarization, rises thus serious questions on the social cohesion of the metropolitan area.
Knowledge of factors inducing social polarization of the municipalities in the metropolitan area is nevertheless uncertain. Several factors contribute to the valorization or to the devalorization of the municipal residential space. But these factors have “soft”, uncertain impacts on the phenomena under investigation : the same causes can sometimes produce different effects. A probabilistic model of these socio-spatial mechanisms has already been proposed (Scarella 2014) in the form of a Bayesian network (Jensen 2001). More particularly, this model was used to investigate several scenarios of future evolution of spatial polarization in the metropolitan area.

Nevertheless, the uncertainty content of model results has not been completely explored. Moreover, alternative theoretic frameworks exist to model uncertain knowledge. Possibility theory (Dubois and Prade 1988, 2001) seems particularly appropriate to model epistemic uncertainties in the knowledge of geographic phenomena and of the causal relations among them. Thanks to recent advancements in the implementation of possibilistic networks (Caglioni et al. 2014), a new possibilistic model has thus been developed (Dubois et al. 2015).
Both the probabilistic and the possibilistic model are used to produce trend scenarios for social polarization in the metropolitan area of Marseille. Both scenarios are based on uncertain knowledge of relationships among variables and produce uncertain evaluation of the future state of the metropolitan area in terms of social polarization. The Tableau® dataviz platform was used to create an interactive on-line geo-visualization of the two scenarios. A particular attention was given to the visualization of the uncertain content of the two trend scenarios. Only appropriate visualizations can indeed make uncertain knowledge useful for scientific understanding and for planning policies (Harrower 2003). The user can appreciate how the visualizations are modified when different levels of certainty/uncertainty are requested for the model results.


Figure 1 - Probabilistic and possibilistic trend scenarios of the social polarization in the metropolitan area of Marseille.

 

Geo-dataviz :
https://public.tableau.com/profile/fusco#!/vizhome/RepresentingUncertainFutures/Story1

 

References :

  • Caglioni M., Dubois D., Fusco G., Moreno D., Prade H., Scarella F., Tettamanzi A., 2014, Mise en œuvre pratique de réseaux possibilistes, LFA 2014, Cargèse 22-24 novembre 2014, p. 267-274
  • Centi C. (1993), Les enjeux du modèle niçois. L’approche localiste du développement en question, Revue Economique, vol 44, n. 4, pp.685-712
  • Centi C. (1996), Le laboratoire marseillais : chemins d’intégration métropolitaine et de segmentation sociale, Paris, L’Harmattan.
  • Dubois D., Fusco G., Prade H., Tettamanzi A., 2015, Uncertain Logical Gates in Possibilistic Networks. An Application to Human Geography. In Ch. Beierle and A. Dekhtyar (Eds.). Scalable Uncertainty Management - 9th International Conference, SUM 2015, Québec City, QC, Canada, September 16–18, 2015. Proceedings, Lecture Notes in Artificial Intelligence, vol. 9310, Springer, pp. 249–263.
  • Dubois D., Prade H. (1988), Possibility Theory : An Approach to Computerized Processing of Uncertainty, Plenum Press.
  • Dubois D., Prade H. (2001), Possibility Theory, Probability Theory and Multiple-valued Logics : A Clarification, Annals of Mathematics and Artificial Intelligence, 32, p. 35–66
  • Fusco G., Scarella F. (2011), Métropolisation et ségrégation sociospatiale. Les flux des migrations résidentielles en PACA, L’Espace Géographique, tome 40, n°4, pp. 319-336
  • Jensen F. (2001), Bayesian Networks and Decision Graphs, New York, Springer.
  • Harrower, M. (2003) Representing Uncertainty : Does it help people make better Decisions, In UCGIS Workshop : Geospatial Visualization and Knowledge Discovery Workshop, Lansdowne, Virginia.
  • Scarella F. (2014), La ségrégation résidentielle dans l’espace-temps métropolitain : analyse spatiale et géo-prospective des dynamiques résidentielles de la métropole azuréenne, Thèse de Doctorat de Géographie, Université de Nice Sophia Antipolis.