Extractive Text Summarization using TextRank algorithm

This project can prepare a summary of given text in desired number of sentences given by user.

Demo

https://summarizerproject.herokuapp.com/

Features

• Paragraph Summarization (max length 10000)
• URL Summarization
• Document Summarization (Upload .pdf or .docx or .txt file max length 10000)
• dictionaryapi json response for searching word
• send generated summary to email
• user account creation, login and django admin dashboard

Run Locally

Clone the project

  git clone https://github.com/awg01/summarize.git

Go to the project directory

  cd summarize

Download glove.6B.100d.txt(347.12 MB) file from https://www.kaggle.com/datasets/rtatman/glove-global-vectors-for-word-representation and add inside project folder. Same folder where manage.py file resides.

Install dependencies

  pip install -r requirements.txt

Start the server

  python3 manage.py runserver

Deployment

To deploy this project on heroku read this article

https://codeburst.io/deploy-your-django-project-for-free-140d73a2c76b

Algorithm Implementation

Google collab notebooks

• sumy python library demo
  • https://colab.research.google.com/drive/15csC9Y3kANaPRYCAcSclU-VnsRlhXO1m?usp=sharing
• TextRank algorithm implementation
  • https://colab.research.google.com/drive/18M63bnEQv42GN3hZIMmgxXVMJ5dgOBFm?usp=sharing

Appendix

Any additional information goes here

https://www.machinelearningplus.com/nlp/text-summarization-approaches-nlp-example/

https://www.analyticsvidhya.com/blog/2018/11/introduction-text-summarization-textrank-python/

https://medium.com/data-science-in-your-pocket/text-summarization-using-textrank-in-nlp-4bce52c5b390

My twitter profile id

https://twitter.com/AkashGudadhe4