sentopicmodel: Create a sentopic model

In sentopics: Tools for Joint Sentiment and Topic Analysis of Textual Data

Create a sentopic model

Description

The set of functions LDA(), JST(), rJST() and sentopicmodel() are all wrappers to an unified C++ routine and attempt to replicate their corresponding model. This function is the lower level wrapper to the C++ routine.

Usage

sentopicmodel(
  x,
  lexicon = NULL,
  L1 = 5,
  L2 = 3,
  L1prior = 1,
  L2prior = 5,
  beta = 0.01,
  L1cycle = 0,
  L2cycle = 0,
  reversed = TRUE
)

Arguments

x	tokens object containing the texts. A coercion will be attempted if x is not a tokens.
lexicon	a quanteda dictionary with positive and negative categories
L1	the number of labels in the first document mixture layer
L2	the number of labels in the second document mixture layer
L1prior	the first layer hyperparameter of document mixtures
L2prior	the second layer hyperparameter of document mixtures
beta	the hyperparameter of vocabulary distribution
L1cycle	integer specifying the cycle size between two updates of the hyperparameter L1prior
L2cycle	integer specifying the cycle size between two updates of the hyperparameter L2prior
reversed	indicates on which dimension should lexicon apply. When reversed=FALSE, the lexicon is applied on the first layer of the document mixture (as in a JST model). When reversed=TRUE, the lexicon is applied to the second layer of the document mixture (as in a reversed-JST model).

Value

An S3 list containing the model parameter and the estimated mixture. This object corresponds to a Gibbs sampler estimator with zero iterations. The MCMC can be iterated using the fit() function.

• tokens is the tokens object used to create the model

• vocabulary contains the set of words of the corpus

• it tracks the number of Gibbs sampling iterations

• za is the list of topic assignment, aligned to the tokens object with padding removed

• logLikelihood returns the measured log-likelihood at each iteration, with a breakdown of the likelihood into hierarchical components as attribute

The topWords() function easily extract the most probables words of each topic/sentiment.

Author(s)

Olivier Delmarcelle

See Also

Fitting a model: fit(), extracting top words: topWords()

Other topic models: JST(), LDA(), rJST()

Examples

LDA(ECB_press_conferences_tokens)
rJST(ECB_press_conferences_tokens, lexicon = LoughranMcDonald)

sentopics documentation built on Sept. 20, 2024, 5:06 p.m.