Search the site

Home » Spatial modelling park » Spatial analysis tools

GéoGrapheur: Exploratory Data Analysis Module

Author of the model: Arnaud BANOS, Géographie-cités, Paris, France

Creation date:
- initial: 2001
- of this version: 2001

GéoGrapheur is an exploratory spatial data analysis module that aims to be ergonomic, interactive and dynamic.

In its current state, only vectorial information is treated (the integration of a model allowing the sowings of points is in progress).

GéoGrapheur draws part of the graphic and statistical capacities from the software ARC (Cook and Weisberg), as well as the capacities of spatial information management provided by Livemap (Brunsdon). On this basis, three families of spatial analysis methods are proposed:

1) A spatial smoothing module, which contains traditional methods such as the moving average (weighted or not), but also other stronger methods: mobile medians, doubled mobile averages and medians, iterative mobile medians.

2) A spatial autocorrelation module (global and local). The coefficients of Moran and Geary were initially favoured. A test of significance based on a Monte Carlo simulation is also proposed.

3) A geographical local regression module, offering a local spatial extension of any statistical model belonging to the family of the generalized linear model. In addition, a dynamic interface allows the visualisation of the "spatial moving" of the selected function.

Theoretical references:

The ambition of GéoGrapheur is to contribute to the diffusion of three more or less linked great intellectual positions:

  • exploratory data analysis (EDA), initiated by the American mathematician John Wilder Tukey
  • exploratory spatial data analysis (ESDA)
  • the Anglo-Saxon "Geocomputation" movement

GéoGrapheur aims to associate these three movements within a single environment, while avoiding becoming a patchwork of methodologies.

Inputs
- Values, tables
- Maps

Outputs
- Values, tables
- Maps

Zip - 5 Mb
Download Geographeur

Comments

This model is a "space" extension of the software ARC. Some rules must however be complied with to support the authors in "the collective adventure of free software": any use of XLISPSTAT must quote the authors (Betz, Tierney), idem for all the affiliated software: ARC (Cook Weisberg), VISTA (Young), LiveMap (Brunsdon), ARPEGE’ (Josselin), MIASME (Banos)...

It is formally forbidden to take the code and to use it for any commercial purpose. No support nor maintenance for individual users is provided for the use of this software, all comments are welcome and should be addressed to the authors.

Bibliographic references

The article furnished with the software and titled "A propos de l’analyse exploratoire de données" sets out the epistemological setting of GéoGrapheur.

Additional information can be found in:

  • BANOS A., 1999 : "Quelle implication de l’utilisateur dans une stratégie de data mining spatial ? Illustration à partir de l’appréhension spatio-temporelle des accidents de la route en milieu urbain", Revue Internationale de Géomatique, Vol. 9, n° 4, pp. 441-456
  • BANOS A., HUGUENIN-RICHARD F., 1999, "Spatial distribution of road accidents in the vicinity of point sources. Application to child pedestrian accidents", Geography and Medicine, Elsevier, pp. 54-64

 Key word(s) linked to this article

Software » download



This article last updated Wednesday 1 April 2009. by Richard Stephenson