The Research of the Data Mining Based on the Spatial Database Technology

With the wide application of GIS to all kinds of fields, and developing of the technique of data mining and spatial data collection, the technique of data mining in spatial database-spatial data mining is coming out. In order to satisfy the people´s demand for the interesting and potentially useful knowledge from the spatial database, this thesis used a wide using spatial clustering algorithm: k-means algorithm to discover interesting and potentially useful spatial patterns embedded in spatial database, and also has realized an improved genetic algorithm based on the k-means algorithm. The improved genetic algorithm not only have the global search advantage of genetic algorithm, but also have the feature of local convergence fast of k-means algorithm, meanwhile, it overcome the sensitive to the initial election data and easily fall into the local optimal drawback by traditional k-means algorithm and also raise the convergence rate.