Research On Spatial Query Language And Spatial Data Mining Based On GIS

Fast growing of spatial information and the complexity is far beyond the capacity of people's interpretation. Although spatial database enables data input, changes, statistics, queries and other functions, but can not effectively find out hidden in the back of the relations between spatial data, rules and trends and other features, which results situation of "Spatial data rich, space knowledge poor". Therefore, from the spatial database automatically mining knowledge, to find not clear, implicit knowledge, spatial relations, or other modes becomes increasingly important. Content of this thesis started at spatial query language, and then deeply researched on space online analytical processing and spatial association rules.According to the characteristics of spatial data, this thesis proposed a new spatial query language, Semantic Query Description Language for Geography (SQDL-G), based on GIS. The language converts spatial predicates into spatial operators and introduces sub-query structure. The language can express complex spatial query expression, and have a better efficiency than traditional query language. With the support of spatial query language, this thesis studied on the initial stage of spatial data mining-Space online analytical processing (SOLAP). Cooperating with spatial query language (SQDL-G), SOLAP system introduces factor objects and surrounding things into data cubic to setup main spatial dimensions and auxiliary spatial dimensions. Multi-spatial-dimension SOLAP makes full use of topology ability of GIS, materializes spatial relation between objects, and enhances analysis ability of SOLAP. At same time, this thesis did corresponding research in one direction of the advanced stage of spatial data mining-spatial association rules. A new discovering spatial association rules method was proposed, which base on co-location pattern and buffer analysis. This method needn't to manually select a center object, and suit for all kinds of spatial objects.Overall, this thesis made a useful spatial data mining exploration and research. Rich semantics to express a variety of spatial query language SQDL-G, with the construction of multi-dimensional data cube of space, and to participate in the association rules based on analysis of the buffer, making spatial data mining and knowledge discovery become standardized and engineering.