Mathematical Sciences: Clustering Algorithms

数学科学：聚类算法

• 批准号: 8617919
• 负责人: John Hartigan
• 金额: $ 25.76万
• 依托单位: Yale University
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 1987
• 资助国家: 美国
• 起止时间: 1987-08-15 至 1991-01-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=8617919&HistoricalAwards=false
• 关键词: Mathematical; Sciences; Clustering; Algorithms

The research will deal with the broad area of cluster analysis. Cluster analysis is classification of objects in to groups of similar ones. The principal investigator has been a pioneer in the field of cluster analysis. He will investigate several statistical problems connected with clustering. The major problems with the clustering algorithms are to answer the following four questions. (1) How many clusters? (2) Are the clusters real? (3) Are the clusters reliable? and (4) What algorithms should be used. Statistical methods will help answer these questions. Multivariate density estimation procedures will be used. Some of these techniques are used in spatial analysis as well.

这项研究将涉及聚类分析的广泛领域。聚类分析是将对象分类为相似对象的组。首席研究员一直是聚类分析领域的先驱。他将研究几个与集群相关的统计问题。聚类算法的主要问题是回答以下四个问题。(1)有多少簇？(2)簇是真实的吗？(3)簇可靠吗？以及(4)应该使用什么算法。统计方法将有助于回答这些问题。将使用多变量密度估计程序。其中一些技术也用于空间分析。