CleverTind interacts with the Tinder api to like and download users profiles. It starts and holds conversations with matches via cleverbot.

Composite profile pictures can be created from downloaded user profiles using the included average.py script.

The composite creator uses openCV and Haar Cascades to recognize faces. The default training data is included here, but you can get different cascade sets here or even train your own.

The sample_data directory contains some analyized data from more than 60,000 profiles which I have collected. You can generate your own analysis by running the bot and running through the profiles with profile_parse.py.

The figures directory contains scripts for creating visualizations of the parsed data.

The bot requires pyro, cleverbot, and httparty. The image averager requires python 2.7, numpy, and openCV (for facial detection) The quick image averager (no facial detection) requires PIL

site_gen.rb can be used to generate a bunch of markdown files for messages history. It will place the files inside the site directory where there are some files for Pelican to generate a site. You can generate the pages and then the site as follows:

ruby site_gen.rb
cd site
make html
make serve

This will put a server on localhost:8000 where you can view the site. See the Pelican Docs for more info on using Pelican.