期刊
BIOMETRICS
卷 77, 期 4, 页码 1215-1226出版社
WILEY
DOI: 10.1111/biom.13376
关键词
cluster analysis; high-dimension low-sample size; hypothesis testing; unimodality; unsupervised learning
资金
- National Institute of Dental and Craniofacial Research [R03DE023592]
- National Institute of Environmental Health Sciences [P03ES010126]
- National Center for AdvancingTranslational Sciences [UL1RR025747]
- National Science Foundation [DGE-1144081]
This paper proposes a novel method to evaluate the significance of identified clusters by comparing the explained variation due to clustering from the original data to a unimodal reference distribution that preserves the covariance structure in the data. The approach is adapted for high-dimension low-sample size settings and can be used to test the null hypothesis and determine the optimal number of clusters.
Cluster analysis is an unsupervised learning strategy that is exceptionally useful for identifying homogeneous subgroups of observations in data sets of unknown structure. However, it is challenging to determine if the identified clusters represent truly distinct subgroups rather than noise. Existing approaches for addressing this problem tend to define clusters based on distributional assumptions, ignore the inherent correlation structure in the data, or are not suited for high-dimension low-sample size (HDLSS) settings. In this paper, we propose a novel method to evaluate the significance of identified clusters by comparing the explained variation due to the clustering from the original data to that produced by clustering a unimodal reference distribution that preserves the covariance structure in the data. The reference distribution is generated using kernel density estimation, and thus, does not require that the data follow a particular distribution. By utilizing sparse covariance estimation, the method is adapted for the HDLSS setting. The approach can be used to test the null hypothesis that the data cannot be partitioned into clusters and to determine the optimal number of clusters. Simulation examples, theoretical evaluations, and applications to temporomandibular disorder research and cancer microarray data illustrate the utility of the proposed method.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据