About

This is a text analysis application, capable of performing common NLP tasks through the API and a web dashboard.
It has been created as part of the GSOC 2018 project of "adding Greek support to spaCy.io" and released as open source package - https://github.com/eellak/text-analysis

Text analysis is performed through spaCy.io NLP framework, supported languages are: Greek, English, German, Italian, French, Dutch, Spanish and Portoguese.

The web application we've used is Django.
For text summarization we use gensim's implementation of the TextRank algorithm.
The Greek language support contains a text classifier that we've built with FastText.
It is trained in 20.000 articles labeled in the following categories:

Αθλητισμός, Ελλάδα, Επιστήμη, Κόσμος, Οικονομία, Περιβάλλον, Πολιτική, Τέχνες, Υγεία

and it attempts to classify a greek text provided on one of these categories.

Created by:

Markos Gogoulos (mgogoulos@mist.io): NLP + Backend code
Giannis Daras (daras.giannhs@gmail.com): NLP + Backend code
Giannis Stergiou (jiannst@gmail.com): Frontend code + design