土壤分类的模糊集成算法与应用  

Fuzzy Clustering Integrated Algorithm and Application in Soil Classification

作　　者：宋明娟[1] 张娜[2] 

机构地区：[1]黑龙江科技学院理学院,黑龙江哈尔滨150027 [2]黑龙江科技学院资源与环境工程学院,黑龙江哈尔滨150027

出　　处：《数学的实践与认识》2011年第23期122-126,共5页Mathematics in Practice and Theory

基　　金：黑龙江省教育厅科学技术研究项目(11511347)

摘　　要：土壤是一个多性状的连续体,其分类的首选方法是模糊聚类分析.但是模糊聚类分析中现有的基于模糊等价关系的动态聚类法和模糊c-均值法各有利弊,采用其中一种方法聚类肯定存在不足.为此集成两种聚类方法的优点,避其缺点,提出了用基于模糊等价关系的动态聚类方法和方差分析方法确定聚类数目和初始聚类中心,再用模糊c-均值法决定最终分类结果的集成算法,并将其应用到松花江流域土壤分类中,得到了较为切合实际的分类结果.Summary of soil is a multi-trait continuum, the preferred method of soil classi- fication is fuzzy cluster analysis. However, the fuzzy dynamic clustering analysis based on equivalence relations and fuzzy c- means clustering analysis have thief advantages and disad- vantages, we can＇t use only one of them to cluster. In this paper, we integrate the advantages of both clustering methods, avoid thief shortcomings, propose a fuzzy clustering integrated algorithm that the number of clusters and initial cluster centers is determined by the dy- namic clustering method and analysis of variance, the final classification result is determined by fuzzy c- mean clustering method. We apply this method in the Songhua River basin soil classification, moreover obtain the ideal result.

关 键 词：土壤分类 基于等价关系的模糊聚类 模糊C-均值聚类 方差分析 模糊聚 类的集成算法 

分 类 号：S155[农业科学—土壤学] O159[农业科学—农业基础科学]