Generate a set of words describing each topic from a fitted STM object. Uses a variety of labeling algorithms (see details).
labelTopics(model, topics = NULL, n = 7, frexweight = 0.5)A labelTopics object (list)
matrix of highest probability words
matrix of highest ranking frex words
matrix of highest scoring words by lift
matrix of best words by score
a vector of topic numbers which correspond to the rows
An STM model object.
A vector of numbers indicating the topics to include. Default is all topics.
The desired number of words (per type) used to label each topic. Must be 1 or greater.
A weight used in our approximate FREX scoring algorithm (see details).
Four different types of word weightings are printed with label topics.
Highest Prob: are the words within each topic with the highest probability (inferred directly from topic-word distribution parameter \(\beta\)).
FREX: are the words that are both frequent and exclusive, identifying words
that distinguish topics. This is calculated by taking the harmonic mean of
rank by probability within the topic (frequency) and rank by distribution of
topic given word \(p(z|w=v)\) (exclusivity). In estimating exclusivity we
use a James-Stein type shrinkage estimator of the distribution
\(p(z|w=v)\). More information can be found in the documentation for the
internal function calcfrex and js.estimate.
Score and Lift are measures provided in two other popular text mining
packages. For more information on type Score, see the R package
lda or the internal function calcscore.
For more information on type Lift, see the R package maptpx
or or the internal function calclift.
stm plot.STM
calcfrex js.estimate calcscore calclift