pytypo

corrects English spelling mistakes and normalize. (e.g., "cooooooooooooooollllllllllllll" to "cool")


Keywords
normalize, english-word, nlp, spelling-correction, typo
License
MIT
Install
pip install pytypo==0.3.0

Documentation

pytypo

travis-ci.org coveralls.io Code Health pyversion latest version license

pytypo corrects English spelling mistakes. That feature is based on TYPO CORPUS (http://luululu.com/tweet/) and Wikipedia (https://en.wikipedia.org/wiki/Wikipedia:Lists_of_common_misspellings/For_machines)

And this module normalizes also lengthened English expression having repeating letters. (e.g., this module converts "cooooooooooooooollllllllllllll" to "cool")

That feature is based on the following paper: Samuel Brody and Nicholas Diakopoulos. Cooooooooooooooollllllllllllll!!!!!!!!!!!!!! using word lengthening to detect sentiment in microblogs. In EMNLP2011, pp. 562-570, 2011. http://aclweb.org/anthology//D/D11/D11-1052.pdf

Contributions are welcome!

Installation

$ pip install pytypo

Usage

Import pytypo

>>> import pytypo

correct sentence

>>> pytypo.correct_sentence('you are coooolll!!!')
you are cool!
  • correct_sentence(str)

correct word

>>> pytypo.correct('okayyyyy')
okay
  • correct(str)

Shorten repeated substring until threshould without dictionary

>>> pytypo.cut_repeat('mamisaaaaaan', 1)
mamisan
>>> pytypo.cut_repeat('okayyyyy', 2)
okayy
  • cut_repeat(str, threshould)
    • Note that this method don't use a lengthened expression normalize table (e.g., cooll -> cool). If you want to normalize such expression, use correct() or correct_sentence() method.

License

  • This module is licensed under MIT License.