topic_coherence: A function to calculate topic coherence for a given topic...
In matthewjdenny/SpeedReader: High Performance Text Analysis

A function to calculate topic coherence for a given topic using the formulation in "Optimizing Semantic Coherence in Topic Models" available here: <http://dirichlet.net/pdf/mimno11optimizing.pdf>

1 2	topic_coherence(top_words, document_term_matrix, vocabulary = NULL, numeric_top_words = FALSE, K = length(top_words))

`top_words`	A string vector of top words associated with a topic. If numeric_top_words == TRUE then a numeric vector of word indicies.
`document_term_matrix`	A numeric matrix or data.frame with dimensions number of documents X vocabulary length, where each entry is the count of word j in document i.
`vocabulary`	A string vector containing all words in the vocabulary. The vocaublary vector must have the same number of entries as the number of columns in the document_term_matrix, and the word indicated by entries in the i'th column of document_term_matrix must correspond to the i'th entry in vocabulary. If numeric_top_words == TRUE then it is not necessary to supply.
`numeric_top_words`	Defaults to FALSE. If TRUE, then the function expects a vector of word indicies instead of a string vector of actual words.
`K`	The number of top words to use in calculating the topic coherence. Defaults to the lneght of top_words. Common values are usually in the range of 10-20.