/dating_app

RAMP starting kit on lovoo kisses regression

Primary LanguageHTML

RAMP starting kit on predicting likes on Dating App

Authors: Charlotte Cournon, Solène Debuysère, Godefroy Du Chalard, Yanis Lalou, Agathe Minaro, Luka Trailovic

Our society is evolving, becoming more complex, more intensive and, above all, more numeric. In today’s digital era, social media platforms have become an omnipresent part in our daily life. Sharing, liking, commenting and posting are the basic actions to show our interest in other people's passions or to simply share our own. Very quickly, new types of social networks have emerged, for example to create professional networks, to talk anonymously or to date. In fact, these dating applications are used on a much larger scale than one might think, and therefore bring many challenges with their use.

As dating apps such as Tinder or Lovoo grew in popularity, individuals sought to optimize their profiles to improve their scores of receiving more matches and potential connections.

The objective of this RAMP challenge is to investigate these questions by collecting user profile data and metrics in order to better understand user behaviours and relationships on Lovoo.

Set up

Open a terminal and

  1. install the ramp-workflow library (if not already done)
$ pip install ramp-workflow
  1. install the required libraries for this challenge from the root of the starting kit
$ pip install -U -r requirements.txt

If you are using conda, we provide an environment.yml file, so you can do

$ conda env create -f environment.yml
  1. Follow the ramp-kits instructions from the wiki

Local notebook

Get started on this RAMP with the dedicated notebook.

To test the starting-kit, run

ramp-test --quick-test

Help

Go to the ramp-workflow wiki for more help on the RAMP ecosystem.