Mining Spatial Association Rules in Two-direction
Zuo Wang · 2006
Spatial data mining is different from data mining in transaction DB.In most case,the relationships between the tuples in transaction DB do not be taken into account.But in spatial database,there are relationships not only between the attributes,but also between the tuples,and most of the associations exists between the tuples—objects,such as adjacent,intersection,overlap and other topological relationships.So the tasks of spatial data association rules mining include not only mining the relationships between attributes of spatial objects,which we call vertical direction DM,but also mining the relationships between the tuples,which we call horizontal direction DM.This paper analyses the storage models of spatial data,uses for reference the technologies of data mining in transaction DB,defines spatial association rules,including vertical direction association rule,horizontal direction association rule and two-direction association rule,discusses the measurements of interestingness of spatial association rules,and propose the work flows of spatial association rules data mining.During two-direction spatial association rules mining,we propose an algorithm to get non-spatial itemsets.By spatial analysis,we can transfer the spatial relations into non-spatial associations and get non-spatial itemsets.Based on the non-spatial itemsets,we can make use of Apriori algorithm or other algorithms to get the frequent itemsets and then,spatial association rules come into being.To confirm that,we mine in the land using spatial DB to get spatial association rules to validate the algorithm.The test results show that the algorithm is efficient and can mine the interesting spatial rules.