stm: plot.topicCorr – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

plot.topicCorr

Plot a topic correlation graph

Description

Uses a topic correlation graph estimated by topicCorr and the igraph package to plot a network where nodes are topics and edges indicate a positive correlation.

Usage

## S3 method for class 'topicCorr'
plot(
  x,
  topics = NULL,
  vlabels = NULL,
  layout = NULL,
  vertex.color = "green",
  vertex.label.cex = 0.75,
  vertex.label.color = "black",
  vertex.size = NULL,
  ...
)

Arguments

`x`	A topicCorr model object.
`topics`	A vector of topics to include in the plot, defaults to all.
`vlabels`	A character vector of labels for the vertices. Defaults to "Topic #"
`layout`	The layout algorithm passed to the `igraph` package. It will choose `layout.fruchterman.reingold` by default. Note that to pass an alternate algorithm you should load the `igraph` package first.
`vertex.color`	Color of the vertices.
`vertex.label.cex`	Controls the size of the labels.
`vertex.label.color`	Controls the color of the labels.
`vertex.size`	Controls the sizes of the vertices, either NULL, a scalar or a vector of the same length as number of topics.
`...`	Additional parameters passed to `plot.graph.adjacency`

Details

Essentially a thin wrapper around the plotting functionality in the igraph package. See package vignette for more details.

References

Csardi G, Nepusz T: The igraph software package for complex network research, InterJournal, Complex Systems 1695. 2006. http://igraph.sf.net

Examples

#This function becomes more useful with larger numbers of topics.
#it is demonstrated here with a small model simply to show how the syntax works.
cormat <- topicCorr(gadarianFit)
plot(cormat)

stm

Estimation of the Structural Topic Model

v1.3.6

MIT + file LICENSE

Authors

Margaret Roberts [aut], Brandon Stewart [aut, cre], Dustin Tingley [aut], Kenneth Benoit [ctb]

Initial release