Chapter two is all about using freely available datasets to make recommendations. The math is easy to follow and the python examples help digest all the dense bits. Finishing all the examples, I looked up the movielens data and to my surprise, there is a book dataset of some sort as well. This is great!
I am going to follow up with some quick apps as proof of concepts, first with the movie data and then with the book data. I’ll try to put some twist on these to complement what I’ve learned from the book. BTW- the author has made the code available on his blog.
My tools of choice, however, are any flavor of SQL and Coldfusion, maybe rails. I’ll be using mySQL for now, not sure on programming language yet. If anyone cares, I can provide source when I am done.