A Survey of Popular R Packages for Cluster Analysis

Abstract

Cluster analysis is a set of statistical methods for discovering new group/class structure when exploring datasets. This article reviews the following popular libraries/commands in the R software language for applying different types of cluster analysis: from the stats library, the kmeans and hclust functions; the mclust library; the poLCA library; and the clustMD library. The packages/functions cover a variety of cluster analysis methods for continuous data, categorical data or a collection of the two. The contrasting methods in the different packages are briefly introduced and basic usage of the functions is discussed. The use of the different methods is compared and contrasted and then illustrated on example data. In the discussion, links to information on other available libraries for different clustering methods and extensions beyond basic clustering methods are given. The code for the worked examples in Section 2 is available at http://www.stats.gla.ac.uk/˜nd29c/Software/ClusterReviewCode.R.

Association of Lifestyle and Body Composition on Risk Factors of Cardiometabolic Diseases and Biomarkers in Female Adolescents

“…LS was considered a latent variable, that is, not directly observable, and was evaluated by Latent Class Analysis (LCA) [ 37 ]. With the information from the manifested variables, we fit a statistical model that allowed estimating the probability of a given individual belonging to each of the latent variable categories [ 38 ].…”

Adjustment to Higher Education: A Comparison of Students With and Without Disabilities

“…The analysis classified the entire sample into two clusters (groups); The number of clusters was determined based on the “elbow” method. According to this method, the number of clusters for k-means was chosen by fitting k-means models for a range of consecutive numbers, usually 1 up to some maximum number, and plotting an elbow plot of the total within sum of squares (WSS) value for each number of clusters versus that cluster number ( Flynt and Dean, 2016 ). The value for one cluster was WSS = 3358.3, and the value for two clusters was WSS = 1801.9.…”

“…To better understand the adjustment patterns, we classified the participants into clusters using a k-means cluster analysis with the four adjustment subscales. The number of clusters was determined according to the ‘elbow’ method, in which the number of clusters for k-means was chosen by fitting k-means models for a range of consecutive numbers, usually 1 up to some maximum number, and plotting an elbow plot of the total within sum of squares (WSS) value for each number of clusters versus that cluster number ( Flynt and Dean, 2016 ). In addition, cross-tabulations and chi square tests were used to compare the groups on demographic variables.…”

Promote Biomarker Discovery by Identifying Homogenous Primary Headache Subgroups

“…An example of how the statistical analyses can be conducted includes: (1) subgroup identification across headache categories by means of multidimensional scaling, factor analysis and/or cluster analysis (eg, hierarchical cluster analysis, k -means clustering, latent class analysis, and fuzzy cluster models) [20][21][22] and (2) statistical model fitting, eg, of the biochemical compound concentration as a response variable and the headache category, subgroup and confounding factors (eg, sex and age) as explanatory variables while considering the risk of overfitting. [23][24][25][26] Data dimensionality and complexity should be considered during sample size determination.…”