23 April 2019 in IEEE Transactions on Pattern Analysis and Machine Intelligence
Popular clustering algorithms based on usual distance functions (e.g., the Euclidean distance) often suffer in high dimension, low sample size (HDLSS) situations, where concentration of pairwise distances and violation of neighborhood structure have adverse effects on their performance. In this article, we use a new data-driven dissimilarity measure, called MADD, which takes care of these problems. MADD uses the distance concentration phenomenon to its advantage, and as a result, clustering algorithms based on MADD usually perform well for high dimensional data. We establish it using theoretical as well as numerical studies. We also address the problem of estimating the number of clusters. This is a challenging problem in cluster analysis, and several algorithms are available for it. We show that many of these existing algorithms have superior performance in high dimensions when they are constructed using MADD. We also construct a new estimator based on a penalized version of the Dunn index and prove its consistency in the HDLSS asymptotic regime. Several simulated and real data sets are analyzed to demonstrate the usefulness of MADD for cluster analysis of high dimensional data.
1 August 2016 in Journal of Statistical Planning and Inference
26 May 2017 in Technometrics
Several parametric and nonparametric tests of independence between two random vectors are available in the literature. But, many of them perform poorly for high-dimensional data and are not applicable when the dimension exceeds the sample size. In this article, we propose some tests based on ranks of nearest neighbors, which can be conveniently used in high dimension, low sample size situations. Several simulated and real datasets are analyzed to show the utility of the proposed tests. Codes for implementation of the proposed tests are available as supplementary materials.
13 November 2019 in Machine Learning
28 August 2020 in Statistics and Computing