Data Mining Applications In The Gis Attribute Analysis

It is important that valuable information is drawn from the large databases and more and more demand on the knowledge is required. However, a dilemma, less information based on much data, is fell into. KDD(Knowledge Discovery based on Database) whose core is Data Mining is put forward to , which make it possible mining the hidden information from the large complex database , thus giving new hope to solve the problemOn the other hand, it is difficult for attribute data whose application is on retrieval, query and simple statistics to analyze deeply and mine its hidden mode and rule. The status of the SDSS(Spatial Decision Support System) is not desirable due to lack of much valuable knowledge. Thus the application of Data Mining in analyzing the attributes of GIS is discussed in order to support SDSS. The whole work in this paper is listed as followed:Firstly, this paper set forth on management of Spatial Data and general mining process in Spatial Data Mining;Secondly, the deficiency of GIS Spatial Query on the database and the necessity of Data Ming are referred to and the source data is preprocessed; Thirdly, the single-dimension table is discussed including three parts: ◆the theory of Rough is talked about and is used to analyze the preprocessed data. Also, the data is disposed by programming; ◆The disadvantages of the Apriori algorithm are referred to. Based on the original algorithm, advaced_Apriori algorithm is put forward to by perfecting the Apriori algorithm. Compared with the original algorithm, it increases the speed and decreases the memory space in large; ◆The three methods are compared by two measures, time complexity and space complexity.Fourthly, the multi-dimension table is processed and the Data Mining model is put forward to. The Star Schema is constructed in SQL Server, so the data cube. Data Mining is carrying on the DB Miner;Finally, based on the above, conclusion is drawn and some amendments and further research are looked forward to.