allpair_RandIndex

Runs <code>RandIndex</code> for all pairs of cluster results in its
list input and produces a matrix for use by <code>rand_plot</code>.
Understands replicates within <code>k</code> values.

Clusters longitudinal trajectories over time (can be unequally
spaced, unequal length time series and/or partially overlapping series) on
a common time axis. Performs k-means clustering on a single continuous
variable measured over time, where each mean is defined by a thin plate
spline fit to all points in a cluster. Distance is MSE across trajectory
points to cluster spline. Provides graphs of derived cluster splines,
silhouette plots, and Adjusted Rand Index evaluations of the number
of clusters. Scales well to large data with multicore parallelism available
to speed computation.

George Ostrouchov

clustra

Clustering Longitudinal Trajectories

David Gagnon

Hanna Gerlovin

Chen Wei-Chen

Schmidt Drew

Oak Ridge National Laboratory 

U.S. Department of Veteran's Affairs 

allpair_RandIndex function

<dl><dt>results</dt>
<dd>A list with each element packed internally by the
<code>clustra_rand</code> function with elements:<ul>
<li><code>k</code> - number of clusters</li>
<li><code>rep</code> - replicate number</li>
<li><code>deviance</code> - final deviance</li>
<li><code>group</code> - integer cluster assignments
Note that item order is assumed to be the same across all <code>rep</code> and <code>k</code> but
<code>group</code> numbering need not be same. The algorithm only examines if pairs of
items are in same or different clusters within each <code>results</code> list element.</li>
</ul></dd></dl>

Arguments

allpair_RandIndex: helper for replicated cluster comparison — allpair_RandIndex

<dl>

<dt>results</dt>
<dd>A list with each element packed internally by the
<code>clustra_rand</code> function with elements:<ul>
<li><code>k</code> - number of clusters</li>
<li><code>rep</code> - replicate number</li>
<li><code>deviance</code> - final deviance</li>
<li><code>group</code> - integer cluster assignments
Note that item order is assumed to be the same across all <code>rep</code> and <code>k</code> but
<code>group</code> numbering need not be same. The algorithm only examines if pairs of
items are in same or different clusters within each <code>results</code> list element.</li>
</ul></dd>

</dl>

allpair_RandIndex: allpair_RandIndex: helper for replicated cluster comparison

Description

Usage

Value

Arguments