init.centers

<code>init.centers</code> automatically initializes the centers of the
 clusters before running the Cross-Entropy Clustering algorithm.

Splits data into Gaussian type clusters using the Cross-Entropy
Clustering ('CEC') method. This method allows for the simultaneous use of
various types of Gaussian mixture models, for performing the reduction of
unnecessary clusters, and for discovering new clusters by splitting them.
'CEC' is based on the work of Spurek, P. and Tabor, J. (2014)
<doi:10.1016/j.patcog.2014.03.006>.

Simon Garnier

Cross-Entropy Clustering

Kamieniecki Konrad

Spurek Przemyslaw

init.centers function

<dl><dt>x</dt>
<dd>A numeric matrix of data. Each row corresponds to a distinct
observation; each column corresponds to a distinct variable/dimension. It
must not contain <code>NA</code> values.</dd>
<dt>k</dt>
<dd>An integer indicating the number of cluster centers to initialize.</dd>
<dt>method</dt>
<dd>A character string indicating the initialization method to use. 
It can take the following values:<dl>
 <dt>"kmeans++": </dt>
<dd>the centers are selected using the k-means++ algorithm.</dd></dl></dd> <dt>"random": </dt>
<dd>the centers are randomly selected among the values in
 <code>x</code></dd>
</dl>

Arguments

Cluster Center Initialization — init.centers

<dl>

<dt>x</dt>
<dd>A numeric matrix of data. Each row corresponds to a distinct
observation; each column corresponds to a distinct variable/dimension. It
must not contain <code>NA</code> values.</dd>


<dt>k</dt>
<dd>An integer indicating the number of cluster centers to initialize.</dd>


<dt>method</dt>
<dd>A character string indicating the initialization method to use. 
It can take the following values:<dl>
 <dt>"kmeans++": </dt>
<dd>the centers are selected using the k-means++ algorithm.</dd>

 <dt>"random": </dt>
<dd>the centers are randomly selected among the values in
 <code>x</code></dd>


</dl></dd>

</dl>

init.centers: Cluster Center Initialization

Description

Usage

Value

Arguments

References

Examples