520 likes | 653 Views
Interoperating with GIS and Statistical Environment for an Interactive Spatial Data Mining. Didier Josselin, THEMA, UPRESA 6049 du CNRS, Besançon, GDR CASSINI didier.josselin@univ-fcomte.fr http://thema.univ-fcomte.fr/didier.htm Xlisp-Stat programming :
E N D
Interoperating with GIS and Statistical Environmentfor an Interactive Spatial Data Mining Didier Josselin, THEMA, UPRESA 6049 du CNRS, Besançon, GDR CASSINI didier.josselin@univ-fcomte.fr http://thema.univ-fcomte.fr/didier.htm Xlisp-Statprogramming : @ D. Betz, L. Tierney, C. Brunsdon, D. Josselin, L. Guerre, B. Dancuo
French Research Group about GIS(1990-2000 : GDR CASSINI, 2000...?)
The spatial data mining quest Finding significant relations between geographical objects in order to cluster them
Sub-objectives at geographical entityscale • 1st door : the statistical dependency some entities have common characteristics... • 2nd door : the spatial relation some entities are contiguous, closed from each others… • 3rd door : the combinationof spatial and statistical relation some entities are similar and closed...
Sub-objectives at territory and geographical space scale • 1st door : the spatial cutting out and data aggregation : a succession of deriving ... Analysing spatial repartition, Identifiing gradients, Detecting discontinuities... • 2nd door : the spatial auto-correlation measure Global and local • 3rd door : the identification of geographical composite (heterogeneous) entities
Agricultural flows between French communes Commune A Commune B
Commune aggregate with its key and boundary Commune described by an attribute Commune couple flow What are we looking for ?
+ • Various structured query languages • Existing tools to build clean structured databases • Graphical and mapping functionalities • generally open to other softwares
- • Poor in statistical functions • Rarely integrate Exploratory Data Analysis • Need to write queries rather execute them in a graphic way
+ • Numerous statistical functions • Numerous graphic representations • Ease to select objects on screen • Dynamic link between objects • generally open to development by programming
- • Poor in geographical and semiologic functionalities • Does not integrate structured databases functions • Does not include geometrical or topological models
First methodological choice Adding to a statistical environment some mapping and relational functionalitiesARPEGE’ : a tool to Analyse Robustly in Practice and Explore Geographical Environment (XlispStat)
+ • Dynamic link between multiple objects • Relative fastness to support expert decision making • Facilities to implement relations and triggers between objects • Possibility to focus on many crossed selections
- • Difficult to manage with multiscaling • Users may miss some synthetic statistical indicators or automatic methods • Application must be quite simple (RAM limitations) • Combinatory explosion risk !
Second methodological choice Interoperating with a GIS and a statistical environment softwareLAVSTAT : a dynamic Link between ArcView and XlispSTAT
executing LAVSTAT principles Services, DDE Dynamic link with AVLINK Server connecting XlispStat importing ArcView modifying
+ • Dynamic link between GIS and Statistical software • The whole functionalities access to both systems • Increases the ways to investigate spatial data
- • A screen is not enough to explore data • A few time loss to make interoperating the two softwares • Not already stable (memory conflicts)
A few advices for spatial analysis to take reliant decisions in order to shape the future ...
If you have some objectives to reach with data to explore...
Try to dominate time during anaysis and to be inside learning process ... 4
Bring to light all aspects of your problem by multiple representations 6
… and relations between geographical objects through different scales... 10
… which may be well defined (semantic,topology, structural, functional ...) 11