text2vec (version 0.4.0)

get_tcm: Extract term-co-occurence matrix

Description

This function creates a term-co-occurence matrix from a Corpus object.

Usage

get_tcm(corpus)

Arguments

corpus

HashCorpus or VocabCorpus object. See create_corpus, vectorizers for details.

See Also

create_corpus

Examples

# NOT RUN {
txt = movie_review[['review']][1:1000]
it = itoken(txt, tolower, word_tokenizer)
vocab = create_vocabulary(it)
#remove very common and uncommon words
pruned_vocab = prune_vocabulary(vocab, term_count_min = 10, doc_proportion_max = 0.8,
                                doc_proportion_min = 0.001, max_number_of_terms = 5000)

vectorizer = vocab_vectorizer(pruned_vocab, grow_dtm = FALSE, skip_grams_window = 5L)
it = itoken(txt, tolower, word_tokenizer)
corpus = create_corpus(it, vectorizer)
tcm = get_tcm(corpus)
dim(tcm)
# }