A hybrid spatial data clustering method for site selection: The data driven approach of GIS mining

This article applies customer service to be the research background. Spatial data mining method is proposed to solve site selection of the service center. Firstly, a new data model for recording all the information of customer management is given, which transforms the traditional model-driven strategy to data-oriented method. Secondly, a hybrid spatial clustering method named OETTC–MEANS–CLASA algorithm is proposed. It has the advantages of applying k-means algorithm to reduce the result space and using simulated annealing method (CLASA) as result-searching strategy to find more qualified solutions. On the basis of GIS functions, we design deeper analytical function to take spatial obstacle factors, spatial environmental factors, spatial terrain factors, spatial traffic factors and cost factors into account. The result of the experiment declares that the algorithm does better at the both aspects of perform efficiency and result quality.