Spatial Entity Association Rules Extraction Technology Research

Sptial data mining and knowledge discovery are the foreland of the international geography information science.Being the key technique of sptical data mining and knowledge discovery,sptical association rules is to establish relationship of sptical entities of multi-database and realize effective utilization of multi-source data. In allusion to the characteristic of spatial entity,the paper discussed the distilling means of sptical association rules.The main contents and conclusions included:1.Discussed the character,data frame, spacial granularity and associate-trail of vector entity; summarized the definition and sort of sptical association rules and brang out four distilling methods of sptical association rules; Generalized the technical flow of it.2.Came true the didtilling-method based on spatial topologic relationship;Editted and extended the topologic relationship betweet point and line,point and area,line and line,line and area,area and area;defined the topologic predicationbetween entities as disjoint ,intersect,contain and contained.And applied the conclusions in sptical association rules distilling practice.3.Discussed the qualitative and quantitative distance predication;analysed the qualitative and quantitative calculating method of entities; came true the distilling-method based on spatial distance relationship between point and point,point and area,area and area;4.Studied the three models of spatial direction and brang forward the distilling-methods based on 4-direction-sublate-modal 8- direction-sublate-modal and MBR-modal;analysed the excellence and disadvantage and applicability from figure-affect,veracity of distilling and complexity of arithmetic;5.Discussed the researchful actuality and main technical means of attribute matching;discussed the distilling methods based on attribute name comparability and attribute adopted comparability;came true the distilling method based on attribute name comparability according to the characteristic of vector data;6.The paper applied the conclusions in association-info-distilling of entities adopting the distilling method based on spatial topologic and distance relationship;and distilled the spatial associate rule between traffical entities and accessorial establishment adopting the distilling method based on attribute name comparability according to the characteristic of GPS data.