Plot model-based clustering results: BIC, classification, uncertainty and (for univariate and bivariate data) density.

```
# S3 method for Mclust
plot(x, what = c("BIC", "classification", "uncertainty", "density"),
dimens = NULL, xlab = NULL, ylab = NULL, ylim = NULL,
addEllipses = TRUE, main = FALSE, …)
```

x

Output from `Mclust`

.

what

The type of graph requested:

`"BIC"`

`"classification"`

`"uncertainty"`

`"density"`

By default, all the above graphs are produced. See the description below.

dimens

A vector of length one or two giving the integer dimensions of the
desired coordinate projections for multivariate data in case of
`"classification"`

or `"uncertainty"`

plots.

xlab, ylab

Optional labels for the x-axis and the y-axis.

ylim

Optional limits for the vertical axis of the BIC plot.

addEllipses

A logical indicating whether or not to add ellipses with axes corresponding
to the within-cluster covariances in case of `"classification"`

or
`"uncertainty"`

plots.

main

A logical or `NULL`

indicating whether or not to add a title
to the plot identifying the type of plot drawn.

…

Other graphics parameters.

Model-based clustering plots:

`"BIC"`

=BIC values used for choosing the number of clusters.

`"classification"`

=a plot showing the clustering. For data in more than two dimensions a pairs plot is produced, followed by a coordinate projection plot using specified

`dimens`

.`"uncertainty"`

=a plot of classification uncertainty. For data in more than two dimensions a coordinate projection plot is drawn using specified

`dimens`

.`"density"`

=a plot of estimated density. For two dimensional a contour plot is drawn, while for data in more than two dimensions a matrix of contours for pairs of variables is produced.

For more flexibility in plotting, use `mclust1Dplot`

,
`mclust2Dplot`

, `surfacePlot`

, `coordProj`

, or
`randProj`

.

`Mclust`

,
`plot.mclustBIC`

,
`plot.mclustICL`

,
`mclust1Dplot`

,
`mclust2Dplot`

,
`surfacePlot`

,
`coordProj`

,
`randProj`

.

```
# NOT RUN {
precipMclust <- Mclust(precip)
plot(precipMclust)
faithfulMclust <- Mclust(faithful)
plot(faithfulMclust)
irisMclust <- Mclust(iris[,-5])
plot(irisMclust)
# }
```

Run the code above in your browser using DataCamp Workspace