This is a prototype. Do not use in production. TODO: UNDER CONSTRUCTION!!!
==================================================
A kind of REAME.first file ... but in reality the content are my notes
==================================================
This is the very first version of a Recommendation Engine wrapped into a Web Framework.
The application is web based using Flask. The Recommendation Engine is based on Naïve Bayes algorithm to classify unstructured text.
You can run the application (after installing it as intended with the below section notes)
Linux (venv) $ export FLASK_APP=qnarecom.py MS (venv) $ set FLASK_APP=qnarecom.py
(venv) $ flask run
..register your user ...and test the Q&A hyperlink introducing a text as question or requirement...e.g. ... "Person Unique Identification Number" ... or "Dr. Michael Jellinek" .... and see the the recommended answer. Note: Your Training directory will need more information to give you better results. Take into acount abour 1024 sample request files.
The below initial project notes are from my exercises based on The Flask Mega-Tutorial https://blog.miguelgrinberg.com/post/the-flask-mega-tutorial-part-i-hello-world and the Data Mining theory and algorithms as described in The Ancient Art of the Numerati http://guidetodatamining.com/
Enjoy it and please, let me know any comments to make it better/useful. Thank you.
==================================================
TO DO: app sources download $ git config --global http.proxy http://proxy.mycompany:80 $ git clone https://github.com/daniel-julio-iglesias/qnarecom (...)
==================================================
See notes_qnarecom.txt file inside docs directory.