An Improvement Selection Method of GIS Information Generalization Based on Rough Set

On the basis of the existing GIS information generalization method based on rough set,a computing idea is put forward in view of the particularity of GIS information table.Firstly,we discretize the continuous attributes by unsupervised discretization.Secondly,we add the decision attribute according to the value of conditional attributes.Thirdly,we calculate the importance of conditional attributes by attribute reduction of rough set.Lastly,we generalize the objects by dynamic assessment.The experimental results show that the procedure can solve the discretization efficiency problem of GIS information table;and generalize objects by dynamic assessment not only take into account the spatial information as well as the attribute information of objects,but also keep the main spatial feature of the objects after generalization.