K-Means Clustering: Unsupervised Learning for Recommender Systems

Unsupervised Learning has been called the closest thing we have to "actual" Artificial Intelligence, in the sense of General AI, with K-Means Clustering one of its simplest, but most powerful applications. I am not here to discuss whether those claims are true or not, as I am not an expert nor a philosopher. I will however state, that I am often amazed by how well unsupervised learning techniques, even the most rudimentary, capture patterns in the data that I would expect only people to find. Today we'll apply unsupervised learning on a Dataset I gathered myself. It's a database of professional Magic: The Gathering decks that I crawled from mtgtop8.com, an awesome website if you're into Magic: the Gathering. I scraped the MtgTop8 data from a few years of tournaments, and they're all available to be consumed in this GitHub repository.