Package ‘textplot’

October 11, 2020

Type Package
Title Text Plots
Version 0.1.4
Maintainer Jan Wijffels <jwijffels@[Link]>
Description
Visualise complex relations in texts. This is done by providing functionalities for displaying
text co-occurrence networks, text correlation networks, dependency relation-
ships as well as text clustering.
Feel free to join the effort of providing interesting text visualisations.
License GPL-2

URL [Link]
LazyData true
Imports utils, methods, lattice, stats, Matrix, graphics, [Link]
(>= 1.9.6)
Suggests knitr, udpipe, BTM, igraph, graph, Rgraphviz, qgraph, glasso,
ggplot2, ggraph, ggforce, concaveman
VignetteBuilder knitr
RoxygenNote 7.1.1
NeedsCompilation no
Author Jan Wijffels [aut, cre, cph],
BNOSAC [cph],
Sacha Epskamp [ctb, cph] (code in R/matrix_reduction.R adapted from the
qgraph package version 1.4.0 which is GPL-2 licensed),
Ingo Feinerer and Kurt Hornik [ctb, cph] (partial code in
R/textplot_corlines.R adapted from the tm package version 0.4 which
is GPL-2 licensed)
Repository CRAN
Date/Publication 2020-10-10 22:00:02 UTC

1
2 example_udpipe

R topics documented:
example_btm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
example_udpipe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
[Link] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
textplot_bar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
textplot_bitermclusters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
textplot_cooccurrence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
textplot_correlation_glasso . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
textplot_correlation_lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
textplot_correlation_lines_attrs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
textplot_dependencyparser . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

Index 16

example_btm Example Biterm Topic Model

Description
The object is a BTM topic model created with the BTM package. It was created on a subset of all
CRAN packages, namely package which are part of the NaturalLanguageProcessing and Machine-
Learning task views.
Timepoint of creation was 2020-04-10.

Examples

library(BTM)
data(example_btm, package = 'textplot')
example_btm
str(example_btm)

example_udpipe Example annotation of text using udpipe

Description
The object is a [Link] of the annotation of the text: "UDPipe provides tokenization, tagging,
lemmatization and dependency parsing of raw text"

Examples
data(example_udpipe)
str(example_udpipe)
[Link] 3

[Link] Plot function for a BTM object

Description
Plot biterms as a clustered graph. The graph is constructed by assigning each word to a topic and
within a topic of words biterm frequencies are shown.

Usage
## S3 method for class 'BTM'
plot(
x,
biterms = terms(x, type = "biterms")$biterms,
top_n = 7,
which,
labels = seq_len(x$K),
title = "Biterm topic model",
subtitle = list(),
...
)

Arguments
x an object of class BTM with a biterm topic model
biterms a [Link] with columns term1, term2, topic with all biterms and the topic
these were assigned to. Defaults to the biterms used to construct the model.
top_n integer indicating to limit to displaying the top_n terms for each topic. Defaults
to 7.
which integer vector indicating to display only these topics. See the examples.
labels a character vector of names. Should be of the same length as the number of
topics in the data.
title character string with the title to use in the plot
subtitle character string with the subtitle to use in the plot
... not used

Value
an object of class ggplot

See Also
BTM, textplot_bitermclusters.default
4 [Link]

Examples

library(igraph)
library(BTM)
library(ggraph)
library(ggforce)
library(concaveman)
data(example_btm, package = 'textplot')

model <- example_btm

plot(model, title = "BTM model", top_n = 3)

plot(model, title = "BTM model", top_n = 3, labels = 1:model$K)
plot(model, title = "BTM model", which = 7:15)
plot(model, title = "BTM model", subtitle = "First 5 topics",
which = 1:5, top_n = 10)
plot(model, title = "Biterm topic model", subtitle = "First 8 topics",
which = 1:8, top_n = 7)

topiclabels <- c("Garbage",

"Data Mining", "Gradient descent", "API's",
"Random Forests", "Stat models", "Text Mining / NLP",
"GLM / GAM / Bayesian", "Machine learning", "Variable selection",
"Regularisation techniques", "Optimisation", "Fuzzy logic",
"Classification/Regression trees", "Text frequencies",
"Neural / Deep learning", "Variable selection",
"Text file handling", "Text matching", "Topic modelling")
plot(model, title = "Biterm topic model", subtitle = "some topics",
top_n = 7,
which = c(3, 4, 5, 6, 7, 9, 12, 16, 20),
labels = topiclabels)

library(BTM)
library([Link])
library(udpipe)
## Annotate text with parts of speech tags
data("brussels_reviews", package = "udpipe")
anno <- subset(brussels_reviews, language %in% "nl")
anno <- [Link](doc_id = anno$id, text = anno$feedback, stringsAsFactors = FALSE)
anno <- udpipe(anno, "dutch", trace = 10)
## Get cooccurrences of nouns / adjectives and proper nouns
biterms <- [Link](anno)
biterms <- biterms[, cooccurrence(x = lemma,
relevant = upos %in% c("NOUN", "PROPN", "ADJ"),
skipgram = 2),
by = list(doc_id)]
## Build the BTM model
[Link](123456)
x <- subset(anno, upos %in% c("NOUN", "PROPN", "ADJ"))
x <- x[, c("doc_id", "lemma")]
textplot_bar 5

model <- BTM(x, k = 5, beta = 0.01, iter = 2000, background = TRUE,

biterms = biterms, trace = 100)
plot(model)

textplot_bar Barplot of a frequency table using lattice

Description
Barplot of a frequency table using lattice

Usage
textplot_bar(x, ...)

## Default S3 method:
textplot_bar(
x,
panel = "Effect",
total = sum(x),
top = 40,
[Link] = "lightgrey",
[Link] = "lightblue",
lwd = 3,
cextext = 0.5,
addpct = FALSE,
cexpct = 0.75,
textpos = 3,
pctpos = 1,
v = NULL,
[Link] = "red",
...
)

Arguments
x a table to plot or a [Link] with the first column the label and the second
column the frequency
... other arguments passed on to lattice::dotplot
panel character string what to put into the panel
total integer with the total. Defaults to sum(x). Is used to plot the table counts as a
percentage. In which case this is divided by the total.
top integer indicating to plot only the first ’top’ table elements. Defaults to 40.
6 textplot_bitermclusters

[Link] color of the panel. Defaults to lightgrey.

[Link] color of the line. Passed on to the col argument in lattice::[Link]
lwd width of the line. Passed on to the lwd argument in lattice::[Link]
cextext numeric with the cex of the text with the counts plotted. Passed on to lattice::[Link].
addpct logical indicating to add the percent with lattice::[Link]
cexpct numeric with the cex of the text plotted when using addpct. Passed on to
lattice::[Link].
textpos passed on to the pos argument of [Link] to indicate where to put the text of
the frequencies
pctpos passed on to the pos argument of [Link] to indicate where to put the text of
the percentages
v passed on to lattice::[Link] to draw a vertical line
[Link] passed on to lattice::[Link] to draw a vertical line

Value
the result of a call to lattice::dotplot

Examples

data(brussels_listings, package = 'udpipe')

x <- table(brussels_listings$neighbourhood)
x <- sort(x)
textplot_bar(x,
panel = "Locations", [Link] = "darkgrey", xlab = "Listings",
cextext = 0.75, addpct = TRUE, cexpct = 0.5)

x <- sample(LETTERS, 1000, replace = TRUE)

textplot_bar(sort(table(x)), panel = "Frequencies", xlab = "Frequency",
cextext = 0.75, main = "Freq stats")
textplot_bar(sort(table(x)), panel = "Frequencies", addpct = TRUE, top = 15)

## x can also be a [Link] where the first column

## is the label and the second column the frequency
x <- [Link](l = LETTERS, amount = rnorm(26))
textplot_bar(x)
textplot_bar(x, v = 0)

textplot_bitermclusters
Plot biterm cluster groups
textplot_bitermclusters 7

Description
Plot biterms as a clustered graph. The graph is constructed by assigning each word to a topic and
within a topic of words biterm frequencies are shown.

Usage
textplot_bitermclusters(x, ...)

## Default S3 method:
textplot_bitermclusters(
x,
biterms,
which,
labels = seq_len(length(table(biterms$topic))),
title = "Biterm topic model",
subtitle = list(),
...
)

Arguments
x a list of [Link], each containing the columns token and probability corre-
sponding to how good a token is emitted by a topic. The list index is assumed
to be the topic number
... not used
biterms a [Link] with columns term1, term2, topic with all biterms and the topic
these were assigned to
which integer vector indicating to display only these topics. See the examples.
labels a character vector of names. Should be of the same length as the number of
topics in the data.
title character string with the title to use in the plot
subtitle character string with the subtitle to use in the plot

Value
an object of class ggplot

Examples

library(igraph)
library(ggraph)
library(concaveman)
library(ggplot2)
library(BTM)
data(example_btm, package = 'textplot')
group_terms <- terms(example_btm, top_n = 3)
group_biterms <- example_btm$biterms$biterms
8 textplot_cooccurrence

textplot_bitermclusters(x = group_terms, biterms = group_biterms)

textplot_bitermclusters(x = group_terms, biterms = group_biterms,
title = "BTM model", subtitle = "Topics 7-15",
which = 7:15, labels = seq_len(example_btm$K))

group_terms <- terms(example_btm, top_n = 10)

textplot_bitermclusters(x = group_terms, biterms = group_biterms,
title = "BTM model", subtitle = "Topics 1-5",
which = 1:5, labels = seq_len(example_btm$K))

group_terms <- terms(example_btm, top_n = 7)

topiclabels <- c("Garbage",
"Data Mining", "Gradient descent", "API's",
"Random Forests", "Stat models", "Text Mining / NLP",
"GLM / GAM / Bayesian", "Machine learning", "Variable selection",
"Regularisation techniques", "Optimisation", "Fuzzy logic",
"Classification/Regression trees", "Text frequencies",
"Neural / Deep learning", "Variable selection",
"Text file handling", "Text matching", "Topic modelling")
textplot_bitermclusters(x = group_terms, biterms = group_biterms,
title = "Biterm topic model", subtitle = "some topics",
which = c(3, 4, 5, 6, 7, 9, 12, 16, 20),
labels = topiclabels)

textplot_cooccurrence Plot term cooccurrences as a network

Description
Plot term cooccurrences in a graph structure

Usage
textplot_cooccurrence(x, ...)

## Default S3 method:
textplot_cooccurrence(
x,
terms,
top_n = 50,
title = "Term cooccurrences",
subtitle = list(),
vertex_color = "darkgreen",
edge_color = "grey",
base_family = "",
textplot_correlation_glasso 9

...
)

Arguments
x a [Link] with columns term1, term2 and cooc indicating how many times 2
terms are occurring together
... other parameters passed on to ggraph::geom_node_text
terms a character vector with terms to only plot. Prevails compared to using top_n
top_n integer indicating to show only the top n occurrences as in head(x,n = top_n)
title character string with the title to use in the plot
subtitle character string with the subtitle to use in the plot
vertex_color character with the color of the label of each node. Defaults to darkgreen.
edge_color character with the color of the edges between the nodes. Defaults to grey.
base_family character passed on to theme_void setting the base font family

Value
an object of class ggplot

Examples

library(udpipe)
library(igraph)
library(ggraph)
library(ggplot2)
data(brussels_reviews_anno, package = 'udpipe')
x <- subset(brussels_reviews_anno, xpos %in% "JJ" & language %in% "fr")
x <- cooccurrence(x, group = "doc_id", term = "lemma")

textplot_cooccurrence(x, top_n = 25, subtitle = "showing only top 25")

textplot_cooccurrence(x, top_n = 25, title = "Adjectives",
vertex_color = "orange", edge_color = "black",
fontface = "bold")

textplot_correlation_glasso
Plot sparse term correlations as a graph structure

Description
Plot sparse term correlations as a graph structure. Uses the glasso procedure (glasso::glassopath)
to reduce the correlation matrix to retain only the relevant correlations and next visualises these
sparse correlations.
10 textplot_correlation_glasso

Usage
textplot_correlation_glasso(x, ...)

## Default S3 method:
textplot_correlation_glasso(
x,
n = 1000,
exclude_zero = TRUE,
[Link] = 1,
[Link] = 0.5,
...
)

Arguments
x a correlation matrix
... further arguments passed on to qgraph::qgraph, except layout which is set
to ’spring’, labels (taken from the colnames of x), and borders which is set to
FALSE.
n sample size used in computing the sparse correlation matrix. Defaults to 1000.
exclude_zero logical indicating to exclude zero-correlations from the graph
[Link] passed on to qgraph::qgraph
[Link] passed on to qgraph::qgraph

Value
an object of class ggplot

Examples

library(udpipe)
library(qgraph)
library(glasso)
data(brussels_reviews_anno, package = 'udpipe')
x <- subset(brussels_reviews_anno, xpos %in% "NN" & language %in% "fr" & )
x <- document_term_frequencies(x, document = "doc_id", term = "lemma")
dtm <- document_term_matrix(x)
dtm <- dtm_remove_lowfreq(dtm, maxterms = 60)

m <- dtm_cor(dtm)
textplot_correlation_glasso(m, exclude_zero = TRUE)

textplot_correlation_glasso(m, exclude_zero = FALSE)

textplot_correlation_lines 11

textplot_correlation_lines
Document/Term Correlation Plot

Description

Plots the highest occurring correlations among terms.

This is done by plotting the terms into nodes and the correlations between the terms as lines between
the nodes. Lines of the edges are proportional to the correlation height. This uses the plot function
for graphNEL objects (using the Rgraphviz package)

Usage

textplot_correlation_lines(x, ...)

## Default S3 method:
textplot_correlation_lines(
x,
terms = colnames(x),
threshold = 0.05,
top_n,
attrs = textplot_correlation_lines_attrs(),
terms_highlight,
label = FALSE,
[Link] = 1,
[Link] = "red",
lwd = 1,
...
)

Arguments

x a document-term matrix of class dgCMatrix

... other arguments passed on to plot
terms a character vector with terms present in the columns of x indicating terms to
focus on
threshold a threshold to show only correlations between the terms with absolute values
above this threshold. Defaults to 0.05.
top_n an integer indicating to show only the top top_n correlations. This can be set
to plot only the top correlations. E.g. set it to 20 to show only the top 20
correlations with the highest absolute value.
attrs a list of attributes with graph visualisation elements passed on to the plot func-
tion of an object of class graphNEL. Defaults to textplot_correlation_lines_attrs.
12 textplot_correlation_lines

terms_highlight
a vector of character terms to highlight or a vector of numeric values in the 0-1
range indicating how much (in percentage) to increase the node font size. See
the examples.
label logical indicating to draw the label with the correlation size between the nodes
[Link] cex of the label of the correlation size
[Link] color to use for highlighted terms specified in terms_highlight. Defaults to
red.
lwd numeric value - graphical parameter used to increase the edge thickness which
indicates the correlation strength. Defaults to 1.

Value
invisibly the plot

Examples

## Construct document/frequency/matrix
library(graph)
library(Rgraphviz)
library(udpipe)
data(brussels_reviews_anno, package = 'udpipe')
exclude <- c(32337682L, 27210436L, 26820445L, 37658826L, 33661134L, 48756422L,
23454554L, 30461127L, 23292176L, 32850277L, 30566303L, 21595142L,
20441279L, 38097066L, 28651065L, 29011387L, 37316020L, 22135291L,
40169379L, 38627667L, 29470172L, 24071827L, 40478869L, 36825304L,
21597085L, 21427658L, 7890178L, 32322472L, 39874379L, 32581310L,
43865675L, 31586937L, 32454912L, 34861703L, 31403168L, 35997324L,
29002317L, 33546304L, 47677695L)
dtm <- brussels_reviews_anno
dtm <- subset(dtm, !doc_id %in% exclude)
dtm <- subset(dtm, xpos %in% c("NN") & language == "nl" & )
dtm <- document_term_frequencies(dtm, document = "doc_id", term = "lemma")
dtm <- document_term_matrix(dtm)
dtm <- dtm_remove_lowfreq(dtm, minfreq = 5)
dtm <- dtm_remove_tfidf(dtm, top = 500)

## Plot top 20 correlations, having at least a correlation of 0.01

textplot_correlation_lines(dtm, top_n = 25, threshold = 0.01)

## Plot top 20 correlations

textplot_correlation_lines(dtm, top_n = 25, label = TRUE, lwd = 5)

## Plot top 20 correlations and highlight some terms

textplot_correlation_lines(dtm, top_n = 25, label = TRUE, lwd = 5,
terms_highlight = c("prijs", "privacy"),
main = "Top correlations in topic xyz")

## Plot top 20 correlations and highlight + increase some terms

textplot_correlation_lines(dtm, top_n = 25, label = TRUE, lwd=5,
textplot_correlation_lines_attrs 13

terms_highlight = c(prijs = 0.8, privacy = 0.1),

[Link] = "red")

## Plot correlations between specific terms

w <- dtm_colsums(dtm)
w <- head(sort(w, decreasing = TRUE), 100)
textplot_correlation_lines(dtm, terms = names(w), top_n = 20, label = TRUE)

attrs <- textplot_correlation_lines_attrs()

attrs$node$shape <- "rectangle"
attrs$edge$color <- "steelblue"
textplot_correlation_lines(dtm, top_n = 20, label = TRUE,
attrs = attrs)

textplot_correlation_lines_attrs
Document/Term Correlation Plot graphical attributes

Description

Document/Term Correlation Plot graphical attributes

Usage

textplot_correlation_lines_attrs(fontsize = 25)

Arguments

fontsize size of the font. Defaults to 25

Value

a list with graph visualisation elements used by textplot_correlation_lines

Examples

textplot_correlation_lines_attrs()
14 textplot_dependencyparser

textplot_dependencyparser
Plot output of a dependency parser

Description
Plot output of a dependency parser. This plot takes one sentence and shows for the sentence, the
words, the parts of speech tag and the dependency relationship between the words.

Usage
textplot_dependencyparser(x, ...)

## Default S3 method:
textplot_dependencyparser(
x,
title = "Dependency Parser",
subtitle = "tokenisation, parts of speech tagging & dependency relations",
vertex_color = "darkgreen",
edge_color = "red",
size = 3,
base_family = "",
...
)

Arguments
x a [Link] as returned by a call to udpipe containing 1 sentence
... not used yet
title character string with the title to use in the plot
subtitle character string with the title to use in the plot
vertex_color character with the color of the label of each node. Defaults to darkgreen.
edge_color character with the color of the edges between the nodes. Defaults to red.
size size of the labels in the plot. Defaults to 3.
base_family character passed on to theme_void setting the base font family

Value
an object of class ggplot

See Also
udpipe
textplot_dependencyparser 15

Examples

library(udpipe)
library(ggraph)
library(ggplot2)
library(igraph)

x <- udpipe("The economy is weak but the outlook is bright", "english")

textplot_dependencyparser(x)

x <- udpipe("His speech about marshmallows in New York is utter bullshit", "english")
textplot_dependencyparser(x, size = 4)

x <- udpipe("UDPipe provides tokenization, tagging, lemmatization and

dependency parsing of raw text", "english")
textplot_dependencyparser(x, size = 4)

data("example_udpipe", package = "textplot")

textplot_dependencyparser(example_udpipe, size = 4)
Index

BTM, 3

example_btm, 2
example_udpipe, 2

[Link], 3

textplot_bar, 5
textplot_bitermclusters, 6
textplot_bitermclusters.default, 3
textplot_cooccurrence, 8
textplot_correlation_glasso, 9
textplot_correlation_lines, 11, 13
textplot_correlation_lines_attrs, 11,
13
textplot_dependencyparser, 14

udpipe, 14

16