Website: https://universalturtles.github.io/Lisibilite/

Lisibilite

Repo for maintaining the code base for hybrid readability index calculation application. It is called Lisibilite - meaning Readability in French.

Why this name?

Because, it sounded fun!

Run in Dev Mode

Prereq

1. Miniconda

Install Lisbilite

1. Install Conda

   conda create --name lisibilite python=3

2. Activate Conda

   conda activate lisibilite

3. Install Lisibilite CLI

   pip install -r requirements.txt
   python setup.py install

4. Start Django Server

   export DEBUG="True"
   export DEVELOPMENT_MODE="True"
   cd ui
   python manage.py migrate
   python magage.py runserver

Run the server in server style mode using gunicorn instead of runserver

cd code/ui
gunicorn lisibilite_ui.wsgi:application

Draft

Key Contributors

• Shyam Mohan K
• Viren Chhabria
• Arun Thundyill Saseendran
• Debrup Chakraborty

Tech Stack

Wiki

https://github.com/universalturtles/Lisibilite/wiki

MVP

• A Web UI where a user can paste a block of text and when clicked submit gives back the readability scores

  The readability scores should be easily understandable and a graph possibly Display the text is suitable for what age

• An API that can be access to post a text and gives back a readability score JSON

  Should have token access and should be rate limited

• Users should be able to sign - up and see history of submitted text after sign - in

First Thoughts

• An elegant responsive web application where the user can upload a Doc, PDF or text file and view the readability metrics.
• A user should be able to download a well - formatted report in PDF with a custom headline and the metrics of the uploaded document
• The user should be able to easily understand the interpretation of the metrics with well - formed graphs / charts
• The application should also provide metrics such as the number of words, number of sentences, and number of paragraphs

  The user should have the option to ignore titles, sub - headings, figure captions, table captions, references etc

• There should be support for HATEOAS, HAL compliant fine - grained ReST APIs for all the above features.

  The APIs should be secured by API Tokens(The generation of API tokens for users can be made self - service on a later stage. Initially it will be on request)

Based on Additional Brain Storming

• Word Counts with an ability to filter out articles, titles, figures, tables, references etc., (Academic)
• Ability to create readability reports with custom title and sections Some sections in the report can be
• Lisibilite Readability Graph - Something in my mind
• Word Density Analysis
• Noun Phrase / Verb Phrase Composition Analysis
• And one day - some day - Summary / Abstract of the text
• Text Sentiment - someday - one day
• Spelling Errors
• An API for programmatic access to our service(No one has given it as such)
• Initially, it will be very much rate limited. We can see how we can extend it.
• One day, someday, we can start supporting other languages - that will be yoyo
• I believe we must give the facility of user accounts. We must also categorize some of our features as "Premium", for which they will have to pay: P(Of course all these would be after we get enough user traffic). That would also give us scope for some user - specific analytics.
• The service should be available as MS Word Plugin
• The service should be available as a LibreOffice Plugin
• The service should be available as a responsive cross - platform application server over the cloud.

A sample Django REST Webapp

https: // github.com / rupdeb / PythonWebapp.git