norbertbajko / Telekom-Leading-Data-Hackathon-2016

Prediction for the best hours to receive an advertisement call for a given customer group.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Telekom Leading Data Hackathon - 2016

Authors: Norbert Bajko and Peter Nagy
The application gathered the 1st Price for us in the "Data driven application" category

  • Deep Learning
    We trained a fully connected Nerual Network to predict in which hours of a week is it suitable for a given customer group to receive an advertisement call.

  • Web service
    A simple webservice implemented in Python connecting the network model and the web application.

  • Web client

Image of Webclient

We also imagined a simple use-case for our project for future use. Here you can see our solution integrated to an imaginary CRM, shwowing real-time whether a given customer is available:

Image of UseCase

About

Prediction for the best hours to receive an advertisement call for a given customer group.

License:MIT License


Languages

Language:Python 58.3%Language:HTML 20.4%Language:JavaScript 14.1%Language:CSS 7.3%