Research On Geographical Elements Selection Method Based On Rough Set Theory

The selection method of geographical elements is a key step to map automatic generalization.For those existed geographic elements selection models,selection rules cannot properly integrate various types of attribute information of geographic elements and the model parameters have a certain degree of subjectivity.According to rough set theory,rough sets can data mining classification rules and generate objective decision.This paper attempts to use rough sets to solve the problem of selecting geographical features by exploring the selection rules based on feature attribute data and obtain more objectively selected weights of the index.Firstly,this paper comprehensively discussed the existing research results.By analyzing the advantages and disadvantages of the models and methods,we believe that the geographical element selection model should comprehensively consider the various attributes of the elements,and the selection rules and index weights should be calculated scientifically and effectively.Combined with the development of GIS automatic generalization,this paper points out that it is necessary to build a geographic information feature index library which meets the diversified requirements of cartography in GIS and establish a model for efficient and automatically generating selection decisions.Based on previous research results on automatic generalization and related research on rough set theory,it is concluded that rough sets can be better applied to geographical elements selection.In this paper,by briefly reviewing the rough set theory,a rough set model for geographical feature selection is established.This model transforms the selection of elements into a comprehensive evaluation ranking based on multiple factor indexes,the model extracts the evaluation index sets by using the attribute reduction of rough sets and uses the importance of attributes as the basis for determining the weight of evaluation indicators.In order to study the validity of the model,this paper conducts the feature selection experiments based on rough sets for two geographical factors:residential area and road.By analyzing and discussing the factors that affecting the residential area and road selection,using the rough set model to screen the index set from the research data,establish the necessary indexes and weights for the comprehensive evaluation of residential areas and roads,by obtain the comprehensive selection model and the quantity selection model to get the selection results.Through qualitative analysis and the selection results,the practicability of the geographic element selection model based on rough set theory is verified,the systematic errors of the model are analyzed,and improvement suggestions for the discretization procedure are proposed.