基于三角模糊数的不确定性数据聚类算法  

Research on the clustering algorithm of uncertain data based on triangular fuzzy number

作　　者：陆亿红[1] 翁纯佳 

机构地区：[1]浙江工业大学计算机科学与技术学院,浙江杭州310023

出　　处：《浙江工业大学学报》2016年第4期405-409,共5页Journal of Zhejiang University of Technology

基　　金：水利部公益性行业科研专项(201401044);国家科技支撑计划项目(2012BAD10B01)

摘　　要：随着对实验精确度要求的不断提高,聚类分析中的不确定性数据聚类也越来越受到关注.然而经典的不确定数据聚类通常假设其概率密度函数(PDF)等信息是已知的,而现实过程中,这些指标并没有那么轻易就能获取.考虑到这些情况,可以利用三角模糊数来恰当有效地表示多维不确定性数据,并采用基于三角模糊数的低计算复杂度的距离计算方法,结合K-means基础聚类方法形成一种被命名为UTDK-means(Uncertain triangular fuzzy number data K-means)的聚类方法,而它是基于三角模糊数的.实验结果表明:基于三角模糊数的不确定数据聚类是可行的,具有一定的研究价值.With the increase in the requirements of experimental accuracy,uncertain data clustering method in cluster analysis has more and more attention.Classic uncertain data clustering is generally assumed that the probability density function(PDF)and other information is known,but the reality of the process,these indicators are not so easily able to obtain.In view of this issue,we use triangular fuzzy number to represent the multi-dimensional uncertain data.and the distance calculation method with the low computational complexity based on triangular fuzzy number is combined with K-means method to form a new method called UTDK-means.The experimental results show that the clustering method based on triangular fuzzy number is efficient and worthy to study.

关 键 词：不确定性数据 三角模糊数 聚类算法 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]