Categories &

Functions List

Class Definition: CalinskiHarabaszEvaluation

statistics: CalinskiHarabaszEvaluation

Calinski-Harabasz clustering evaluation.

A CalinskiHarabaszEvaluation object contains the results of evaluating clustering solutions using the Calinski-Harabasz criterion.

The Calinski-Harabasz index (also known as the Variance Ratio Criterion) is determined by the ratio of the between-cluster sum of squares (SSB) to the within-cluster sum of squares (SSW). A higher Calinski-Harabasz index value indicates a better clustering solution, implying that clusters are dense and well-separated.

Create a CalinskiHarabaszEvaluation object by using the evalclusters function with the 'CalinskiHarabasz' criterion.

See also: evalclusters, ClusterCriterion, DaviesBouldinEvaluation, GapEvaluation, SilhouetteEvaluation

Source Code: CalinskiHarabaszEvaluation

The CalinskiHarabaszEvaluation class contains the following properties:

A character vector or a function handle specifying the clustering algorithm used to generate the clustering solutions. It can be empty if the clustering solutions are passed as an input matrix. This property is read-only.

A character vector specifying the name of the criterion used to evaluate the clustering solutions. This property is read-only.

A numeric vector containing the values generated by the evaluation criterion for each clustering solution. This property is read-only.

A numeric vector containing the list of the number of clusters evaluated. This property is read-only.

A logical vector indicating which observations in the data matrix contain missing values (NaN). This property is read-only.

An integer specifying the number of non-missing observations in the data matrix. This property is read-only.

An integer specifying the optimal number of clusters based on the evaluation criterion. This property is read-only.

A numeric vector representing the clustering solution that corresponds to the optimal number of clusters. This property is read-only.

A numeric matrix containing the data used for clustering. This property is read-only.

The CalinskiHarabaszEvaluation class offers the following public methods:

statistics: obj = CalinskiHarabaszEvaluation (x, clust, KList)

obj = CalinskiHarabaszEvaluation (x, clust, KList) clusters the data in x for every cluster count in KList and evaluates each solution. The evaluation runs at construction, so obj arrives with its CriterionValues and OptimalK already set.

  • x is an N×P numeric matrix of observations (rows) and predictors (columns). A row holding a NaN is left out of NumObservations.
  • clust names the clustering method, one of 'kmeans', 'linkage' and 'gmdistribution'; or a function handle that clusters the data; or an N×M numeric matrix of clustering solutions computed elsewhere, one column per cluster count, in which case ClusteringFunction is left empty.
  • KList is a vector of positive integers, the cluster counts to inspect.

evalclusters is the usual way to create one of these objects.

CalinskiHarabaszEvaluation: obj = addK (obj, K)

addK (obj, K) evaluates clustering solutions for the number of clusters specified in the vector K and adds them to the CalinskiHarabaszEvaluation object obj.

See also: CalinskiHarabaszEvaluation, evalclusters

CalinskiHarabaszEvaluation: plot (obj)
CalinskiHarabaszEvaluation: h = plot (obj)

plot (obj) plots the Calinski-Harabasz criterion values against the number of clusters. The optimal number of clusters is marked with an asterisk.

h = plot (obj) additionally returns the handle to the plot axes.

See also: CalinskiHarabaszEvaluation, evalclusters