A new constraint-based spatial clustering algorithm based on spatial adjacent relation for GML data. Issue 1 (1st January 2012)

• Record Type: Journal Article
• Authors: Pattabiraman, V.; Nedunchezhian, R.
• Is Part Of: International journal of knowledge engineering and data mining. Volume 2:Issue 1(2012)
• Page Start: 14
• Page End: 34
• Publication Date: 2012-01-01
• Subjects: GML -- clustering -- spatial clustering -- constraints -- spatial adjacent relations -- spatial data mining; Knowledge representation (Information theory) -- Periodicals; Data mining -- Periodicals; 006.305
• Languages: English
• ISSNs: 1755-2087

Abstract

Spatial data mining (SDM) is the discovery of interesting relationships and characteristics that may exist implicitly in spatial databases. Many methods on spatial clustering have been proposed, but only few of them considered the constraints like road, bridge, tunnel etc., that may present in the spatial data or remotely sensed imagery (RSI) data during the clustering process. The objective of this work is to identify the best constraint-based spatial clustering with help of the spatial adjacent relation and also discover the relationship between spatial and non-spatial attributes. Compared to other spatial clustering algorithms the constraint-based SCAR-GML (CSCAR-GML) clusters the spatial objects in the ideal way based on the spatial adjacent relations. The proposed CSCAR-GML computes the spatial adjacent relations among the spatial objects with the constraints and computing the spatial distance based on the similarity measure of the spatial objects. Then it clusters the spatial objects with the different spatial features according to the computed relations. The objects in one cluster are not nearer to each other, but they have similarity in spatial adjacent relation. The interesting simulation results have been achieved and reported. The experiments show that CSCAR-GML is better for spatial clustering with the constraints.