Program summarizes articles into a few sentences. Inspired by the auto TL;DR bot used for many subreddits
Check it out at http://summarizeme.surge.sh/
Using NLTK
- Remove words with no value from the article (e.g. 'a', 'to', 'am')
- Create frequency table of words
- Assign value to each sentence depending on the words it contains and the frequency table
- Build summary depending on the value of all sentences and the outstanding ones
Flask - Python micro-framework for web development
Angular - Framework for mobile and web applications
Skeleton - CSS Boilerplate
Font Awesome - CSS Toolkit + Icons
virtualenv - Create isolated Python environments
nltk - Natural Language Toolkit
RESTful API hosted in AWS Elastic Beanstalk. Click on URL to check features
Angular application hosted in GitHub Pages using angular-cli-ghpages
Using npm and json-server:
npm install -g json-server
cd client/
json-server --watch db.json
the json server should initialize to http://localhost:3000/, then
npm start
it should redirect you to http://localhost:3001/
To use the RESTful api and get accurate results, set up virtualenv and Flask using pip or your preferred library manager...
cd server
sudo pip install virtualenv
virtualenv venv
. venv/bin/activate * for OS X or Linux *
venv\Scripts\activate * for Windows *
sudo pip install Flask
python restful.py
To deactivate the virtualenv, just type...
deactivate
Check Traversy Media's video: Angular 2 in 60 Minutes
Angular's QuickStart
TL;DR bot made by SMMRY
If interested in understanding how it works, feel free to read my article at dev.to