https://github.com/JohannesBuchner/languagecheck

Improve the language of your paper before submission
https://github.com/JohannesBuchner/languagecheck

grammar-checker language-analysis latex overleaf proof-reading publishing python

Last synced: 2 months ago
JSON representation

Improve the language of your paper before submission

Host: GitHub
URL: https://github.com/JohannesBuchner/languagecheck
Owner: JohannesBuchner
License: bsd-2-clause
Created: 2016-05-19T20:23:29.000Z (over 9 years ago)
Default Branch: master
Last Pushed: 2023-07-24T02:13:53.000Z (over 2 years ago)
Last Synced: 2025-07-16T09:48:24.910Z (6 months ago)
Topics: grammar-checker, language-analysis, latex, overleaf, proof-reading, publishing, python
Language: Python
Size: 4.81 MB
Stars: 106
Watchers: 7
Forks: 14
Open Issues: 0
Metadata Files:
- Readme: README.rst
- License: LICENSE

Awesome Lists containing this project

awesome-scientific-writing - LanguageCheck - Analyses scientific LaTeX papers, suggesting improvements from a list of common mistakes/ambiguities, tense consistency, a vs. an, spell check, and paragraph topic sentences. (Spell Checking and Linting)

README

Language checking for scientific papers
--------------------------------------------

This program attempts to assist you in improving your paper before submission.

Features
---------

* Can analyse any LaTeX papers, and Overleaf projects.
* Makes automated reports to point you to improvements:

* Word level:

* find common grammar mistakes, like wrong prepositions
* find wordy phrases and suggest replacements
* a vs an
* spell-check (using hunspell)

* Sentence level:

* find long, wordy sentences
* check topic sentences

* Paragraph level:

* find tense inconsistencies

* Paper level:

* check visual impression of paper

* All analysis is done offline -- your text does not leave your computer.
* Supports British and American English, but focusses on issues applying to both.

Note that there are false positives -- only you can decide whether a
change would make sense, the reports only point out potential issues.

If you find some rules useless (too many false positives), or you want to add more, please send a pull request!

Demo output
-------------

Example analysis (of an early draft of `this paper `_):

* `Example report for misused phrases `_
* `Overview of all reports `_

Requirements
-------------

* python
* convert command (ImageMagick): Install with your distribution
* nltk: Install with pip
* nltk data: Install with python -m nltk.downloader all
* detex command (usually comes with LaTeX)
* pyhunspell (optional): Install with pip

Installation
--------------

These commands should not give you an error::

$ which convert
$ which python
$ which detex
$ which hunspell
$ ls /usr/share/hunspell/{en_US,en_UK}.{dic,aff}

Then install the python packages and data::

$ pip install pyhunspell --user
$ pip install nltk --user
$ python -m nltk.downloader cmudict stopwords punkt_tab averaged_perceptron_tagger_eng

Usage
--------------

*Using directly*:

* create PDF from your latex file -> mypaper.pdf

* For example, run "pdflatex mypaper.tex"

* use detex to create pure text file -> mypaper.txt

* For example, run "detex mypaper.tex > mypaper.txt". You need detex installed.
* This does not capture figure captions. The detex.sh script can help you include those texts, "bash detex.sh mypaper.tex". You still need detex installed

* run $ python languagecheck.py mydir/mypaper.txt mydir/mypaper.pdf
* open with a web browser mypaper_index.html to see all reports

*Using with Overleaf*::

$ bash languagecheck_overleaf.sh
# for example:
$ bash languagecheck_overleaf.sh https://www.overleaf.com/123456789 mypaper.tex

ecosyste.ms

Data

Tools

Indexes

Applications

Experiments

Awesome

https://github.com/JohannesBuchner/languagecheck

Awesome Lists containing this project

README