Yeah, this article was not well-researched or useful at all. As mentioned upthread, it seems focused on a very specific type of supervised learning that has sin

Yeah, this article was not well-researched or useful at all. As mentioned upthread, it seems focused on a very specific type of supervised learning that has since gone through a major leap in usefulness in the last few years. The problem posited at the beginning of the article is in fact one of the example applications in a couple of ML courses/textbooks.