Prepare the output of an LDA model for use in LDAVis where corpus is created with the quanteda package

topicmodels_json_ldavis.R

#' Convert the output of a topicmodels Latent Dirichlet Allocation to JSON
#' for use with LDAvis
#'
#' @param fitted Output from a topicmodels \code{LDA} model.
#' @param corp Corpus object used to create the document term. 
#' The corpus should be created with the \tm{quanteda} package
#' matrix for the \code{LDA} model. This should have been created with
#' the tm package's \code{Corpus} function.
#' @param doc_term The document term matrix used in the \code{LDA}
#' model. This should have been created with the quanteda package's 
#' \code{convert} function, e.g. \code{convert(twdfm, to="topicmodels")}.
#' where \code{twdfm} is the document-feature matrix created by quanteda's 
#' function \code{dfm}.
#' The code is adapted from https://gist.github.com/christophergandrud/00e7451c16439421b24a
#' 
#' @seealso \link{LDAvis}.
#' @export

topicmodels_json_ldavis <- function(fitted, corp, doc_term){
    # Required packages
    library(topicmodels)
    library(dplyr)
    library(stringi)
    library(tm)
    library(LDAvis)

    # Find required quantities
    phi <- posterior(fitted)$terms %>% as.matrix
    theta <- posterior(fitted)$topics %>% as.matrix
    vocab <- colnames(phi)
    doc_length <- vector()
    for (i in 1:length(corp$documents$texts)) {
        temp <- paste(corp$documents$texts[i], collapse = ' ')
        doc_length <- c(doc_length, stri_count(temp, regex = '\\S+'))
    }
    temp_frequency <- inspect(doc_term)
    freq_matrix <- data.frame(ST = colnames(temp_frequency),
                              Freq = colSums(temp_frequency))
    rm(temp_frequency)

    # Convert to json
    json_lda <- LDAvis::createJSON(phi = phi, theta = theta,
                            vocab = vocab,
                            doc.length = doc_length,
                            term.frequency = freq_matrix$Freq)

    return(json_lda)
}