Skip to content

plashal/ExplainToMe

 
 

Repository files navigation

ExplainToMe

travis licence

Automatic Web Article Summarizer

image

Deploy

What is it?

Explain To Me is a automatic text summarizer, that utilizes TextRank, a graph based algorithm to scans through the contents of a website to extract a concise machine generated summary. The methodology is similar to the way search engines return the most relevant web pages from a users search query.

Support

Here’s a list of Python platforms that are officially supported.

  • Python 2.7
  • Python 3.4
  • Python 3.5
  • pypy 2.5.0 -> 2.7.9

Quickstart

Install

Clone Repository

$ git clone https://github.com/jjangsangy/ExplainToMe.git

Create a virtualenv

$ virtualenv -p python venv

Source Virtualenv

$ source venv/bin/activate

Install Python Dependencies

$ pip install --upgrade pip setuptools wheel
$ pip install -r requirements.txt

Run Server

$ python manage.py runserver
Running on http://127.0.0.1:5000/ (Press CTRL+C to quit)

Now go to your browser and point it towards http://127.0.0.1:5000

Docker

Running ExplainToMe via the official Docker image is an easy way to start a server if you don't want to install python.

We assume here you have already installed Docker for your system.

If you are getting started on OS X, the Docker toolbox is the first thing to checkout.

$ docker run -it -p 5000:5000 jjangsangy/ExplainToMe:latest

Once the server is running, navigate to either localhost:5000 (on Linux) or hostname:5000 (on Mac OS X), where hostname is the IP addresses of your virtual machine, obtained using

$ docker-machine ip my-vm-name

Now access your docker machine ip at port docker-machine-ip:5000

Kitematic

You might also want to try Kitematic on OS X which provides a GUI for running Docker images. Running ExplainToMe through Kitematic is easy, just search for the jjangsangy/ExplainToMe image, start it, and you should see it running

kitematic

Things to look forward to:

  • Summaries of documents in other languages than English!

Releases

No releases published

Packages

No packages published

Languages

  • Python 41.6%
  • HTML 34.9%
  • CSS 21.6%
  • JavaScript 1.9%