Skip to content

Latest commit

 

History

History
88 lines (48 loc) · 2.57 KB

README.md

File metadata and controls

88 lines (48 loc) · 2.57 KB
Raw

NRCLex

(C) 2019 Mark M. Bailey, PhD

About

NRCLex will measure emotional affect from a body of text. Affect dictionary contains approximately 27,000 words, and is based on the National Research Council Canada (NRC) affect lexicon (see link below) and the NLTK library's WordNet synonym sets.

Lexicon source is (C) 2016 National Research Council Canada (NRC) and this package is for research purposes only. Source: http://saifmohammad.com/WebPages/NRC-Emotion-Lexicon.htm As per the terms of use of the NRC Emotion Lexicon, if you use the lexicon or any derivative from it, cite this paper: Crowdsourcing a Word-Emotion Association Lexicon, Saif Mohammad and Peter Turney, Computational Intelligence, 29 (3), 436-465, 2013.

NLTK data is (C) 2019, NLTK Project. Source: [NLTK] (https://www.nltk.org/). Reference: Bird, Steven, Edward Loper and Ewan Klein (2009), Natural Language Processing with Python. O’Reilly Media Inc.

Update

  • Finally got around to cleaning this up a bit. Updated PyPI package with current version. Thanks to all the contributors for cleaning up my terrible code!
  • Expanded NRC lexicon from approximately 10,000 words to 27,000 based on WordNet synonyms.
  • Minor bug fixes.
  • Contributor updated NTC library.

Installation

pip install NRCLex

Affects

Emotional affects measured include the following:

  • fear
  • anger
  • anticipation
  • trust
  • surprise
  • positive
  • negative
  • sadness
  • disgust
  • joy

Sample Usage

from nrclex import NRCLex

#Instantiate NRCLex object, you can pass your own dictionary filename in json format.

text_object = NRCLex(lexicon_file='nrc_en.json')

#You can pass your raw text to this method(for best results, 'text' should be unicode).

text_object.load_raw_text(text: str)

#You can pass your already tokenized text as a list of tokens, if you want to use an already tokenized input. This usage assumes that the text is correctly tokenized and does not make use of TextBlob.

text_object.load_token_list(list_of_tokens: list)

#Return words list.

text_object.words

#Return sentences list.

text_object.sentences

#Return affect list.

text_object.affect_list

#Return affect dictionary.

text_object.affect_dict

#Return raw emotional counts.

text_object.raw_emotion_scores

#Return highest emotions.

text_object.top_emotions

#Return affect frequencies.

text_object.affect_frequencies