heatmap_words: Heatmap of word frequencies by cluster

In deepMOU: Clustering of Short Texts by Mixture of Unigrams and Its Deep Extensions

Description

Displays the heatmap of the cluster frequency distributions of the most frequent terms sorted by the most informative ones.

Usage

heatmap_words(
  x,
  clusters,
  n_words = 50,
  legend_position = "bottom",
  font_size = 12,
  low_color = "grey92",
  top_color = "red",
  main = "Row frequencies of terms distribution",
  xlabel = NULL,
  ylabel = NULL,
  legend_title = "Entropy"
)

Arguments

x	Document-term matrix describing the frequency of terms that occur in a collection of documents. Rows correspond to documents in the collection and columns correspond to terms.
clusters	Integer vector of length of the number of cases, which indicates a clustering. The clusters have to be numbered from 1 to the number of clusters.
n_words	Number of words to include in the heatmap (default is 50).
legend_position	Position of the legend ("none", "left", "right", "bottom", "top", or two-element numeric vector as in theme). Default is "bottom".
font_size	Text size in pts (default is 12).
low_color	Base color for terms with no occurrence in a cluster (default is "grey92")
top_color	Base color for terms concentrated in a single cluster (default is "red")
main	A title for the plot. Default is "Row frequencies of terms distribution".
xlabel	A title for the x-axis. Default is NULL.
ylabel	A title for the y-axis. Default is NULL.
legend_title	A title for the legend. Default is "Entropy".

Details

Takes as input the bag-of-words matrix and returns a heatmap displaying the row frequency distribution of terms according to the clusters. Words are sorted by entropy.

Value

A graphical aid to describe the clusters according to the most informative words.

Examples

# Load the CNAE2 dataset
data("CNAE2")

# Get document labels by clustering using mou_EM
mou_CNAE2 = mou_EM(x = CNAE2, k = 2)

# Usage of the function
heatmap_words(x = mou_CNAE2$x, clusters = mou_CNAE2$clusters)

deepMOU documentation built on March 4, 2021, 9:09 a.m.