制图数据的模糊分级处理模型研究

传统的制图数据分级方法存在对原始数据信息的歪曲、普适性不强及计算复杂等问题。基于此,结合现实分级问题的模糊性,提出基于模糊统计分析模型的制图数据分级处理方法。首先通过专家系统获取各模糊样本集,利用统计分析方法求得样本分布函数;然后利用分布函数获得模糊隶属函数,进而求取各模糊集的最模糊点;最后根据最模糊点获得各模糊集的区域划分,从而实现对制图数据的分级处理。该方法不需要对影响级别划分的多因子进行分析和转换,降低了计算的复杂度;另外,该方法是在获得原始数据实际分布的基础上进行的,在后续的分级过程中避免了对原始数据信息的歪曲。

Research on Fuzzy Classification Model for Mapping Data

It suffers from some server problems when the traditional classification methods have been adopted to classify the real data.First,the traditional methods lead to the misconstruction of the original data.Second,the computation is very complicated.Third,the traditional methods can't be used universally.A modified fuzzy classification method for mapping data has been put forward in this paper.Firstly,achieve the number of classification.Secondly,gain the range of data and the fuzzy sample set by expert system,and compute sample distribution function by statistical analysis.Thirdly,transform the distribution function to fuzzy membership function ,and work out the fuzziest point through the fuzzy membership function.Finally,compute the fuzziest point according to the function and achieve the classified mapping data in according to the fuzziest point of each fuzzy sample set.The calculation complexity of classification is greatly reduced and the proposed method avoids distorting the original information of data.On the one hand,it doesn't need to analyze and convert the factors that have effects on the classification in the proposed method.The fuzzy sample sets are got according to the expert knowledge and the apriori knowledge is made full use,thus the proposed method in this paper is more universal and the result of classification is more impersonal than the traditional classification methods.On the other hand,the fuzzy membership functions have been gained based on the real data distribution,thereby avoiding the distortion of the data and making the classification more reasonable.