Spatial Data Mining Method Based On Formal Concept

In recent years, with the rapid development of the computertechnology and space remote sensing technology, the ability of people to obtain spatialdata is greatly enhanced, and accumulated a large amount of spatial data which hasbecome a valuable untapped resources.So people needed a technology processing tomanagement,analysis and use the spatial data which has been accumulated, and derivethe decision support needs of the people, such as the solution to the problem of globaldesertification and global warming are a large number of spatial data processing andanalysis.Formal Concept Analysis, also called concept lattice analysis, is proposed in the1980s by the German Professor Rudolf Wille, its range of applications, includingmathematical logic, software engineering, data mining, artificial intelligence, andinformation extraction in many fields. Formal concept has been widely used in the field ofdata mining, but has just started in the spatial data mining.In order to solve the spatial data mining association rules extraction problem, throughthe application form concept theory, spatial data in GIS (Geographic Information System)concept lattices constructed,"compact dependent group as a means of data mining, theformation of spatial data and concept lattice analysis process to unify the combination ofangles, from the concept lattice and spatial data mining, spatial data conversion conceptlattice, to find out all the all association rules of the value1of the degree of confidence byusing the compact dependence as a tool, to thereby obtain the space required for dataassociation rules.The main work is as follows:(1) This article describes the spatial data mining to the current domestic andinternational developments, and do a simple introduction to data mining, GIS, spatialtopological predicates.(2) A spatial join index method based on the spatial topological relations wasproposed after analysis spatial objects’ topological,which can be generated on the basis ofthe spatial database formal concept background.(3) This article was first proposed a compact dependent reduction algorithm basic onApriori, in the use of this method mining excavation topological relations between spatialobjects, in order to identify potential implicit in the spatial data spatial topologicalrelations quickly,and verified by specific instance of GIS.