This is very cool. Back when Facebook messenger used xmpp, i wrote a python client to track when friends would sign on and sign off. I built a webapp for my friends to learn about their sleeping habits. I never thought of saving the conversations from eg Pidgin to build a friend recommendation engine.

What is clearly evident is that all the major social media platforms are actively performing these calculations (and more) across their entire user base

What are parasite words, and what exactly does this tool do? I read the readme but couldn't figure it out.

This library will help you find the words parasites that your friend often uses in Telegram. BibaAndBoba can analyze the messages history, and even determine the probability that two people communicate with each other.

If you are interested in how it works, please visit my GitHub repository: https://github.com/andylvua/BibaAndBoba

I would be really delighted if you would leave a star on my GitHub. It's my inspiration.

Whatever "parasite words" are.