sentimentr measures sentence-level sentiment while accounting for negation, intensifiers, downtoners and grouped polarity scores.
Read more
sentimentr measures sentence-level sentiment while accounting for negation, intensifiers, downtoners and grouped polarity scores.
Read more
topicmodels fits latent Dirichlet allocation and correlated topic models, then explores document-topic probabilities in R.
Read more
tokenizers converts text into words, sentences, n-grams, paragraphs and other token types with consistent functions for R analysis.
Read more
textclean prepares noisy text for analysis with replacements for contractions, emoji, URLs, HTML, whitespace and other patterns.
Read more
Ucto splits multilingual text into tokens and sentences, handles punctuation and normalisation, and supports FoLiA annotated output.
Read more
tidytext brings text mining into tidy R workflows with tokenisation, sentiment lexicons, term frequencies, and data visualisation.
Read more
quanteda provides powerful quantitative text analysis with corpus management, tokenisation, document-feature matrices, and dictionaries.
Read more
tm offers an extensible R text-mining framework with document corpora, preprocessing tools, term-document matrices, and metadata.
Read more
stringi delivers fast Unicode string processing with regular expressions, collation, transliteration, encoding and text normalisation.
Read more
stringr simplifies string manipulation in R with consistent functions for matching, extracting, replacing, splitting and joining text.
Read more
UDPipe provides multilingual tokenisation, part-of-speech tagging, lemmatisation, dependency parsing, and keyword extraction in R.
Read more
Word Vectors trains and explores word embeddings in R with word2vec models, nearest-neighbour searches and cosine similarity.
Read more
spacyr connects R to spaCy for tokenisation, part-of-speech tagging, lemmatisation, named entity recognition and dependency parsing.
Read more
MITIE provides named-entity recognition, entity extraction and relation detection with trainable models and multilingual interfaces.
Read more
text2vec supports efficient text vectorisation, document-term matrices, word embeddings, similarity analysis, and topic modelling in R.
Read more
Moses trains and decodes statistical machine translation models from parallel text, with phrase-based, hierarchical, and factored models.
Read more
TiMBL implements memory-based classifiers for NLP, including weighted nearest-neighbour learning, decision trees, and model evaluation.
Read more
MeTA combines text tokenization, indexing, classification, ranking, topic models and linguistic processing in one research toolkit.
Read more
CRF++ provides trainable conditional random fields for sequence tagging, supporting feature templates, n-best output and probabilities.
Read more
BLLIP Parser analyses English sentence structure using a generative constituency parser and discriminative reranking models.
Read more