论文部分内容阅读
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.
Soil is a continuum of many traits, and its preferred method of classification is fuzzy clustering analysis.But the existing fuzzy clustering analysis based on fuzzy equivalent relationship between dynamic clustering and fuzzy c-means have their own advantages and disadvantages, using One of the methods of clustering is definitely insufficient.Therefore, the advantages of the two clustering methods are integrated and the shortcomings of the two clustering methods are integrated.A dynamic clustering method based on fuzzy equivalence relation and variance analysis method are proposed to determine the number of clusters and the initial clustering Center, and then use the fuzzy c-mean method to determine the final classification results of the integrated algorithm, and apply it to the soil classification of the Songhua River Basin, obtained a more realistic classification results.