If you like how the Amazon website recommends products that are perfectly suited to the buyer’s current shopping cart, you’ll love this post.
Here’s a tutorial you can use, with code, that lets you use the Sagemaker service on AWS to create a related products recommendation, adjustable to allow any number of results.
A factorization machine is a general-purpose supervised learning algorithm that you can use for both classification and regression tasks. This algorithm was designed as an engine for recommendation systems. It extends the collaborative filtering approach by learning a quadratic function over the features while restricting second order coefficients to a low rank structure. This restriction is well-suited for large and sparse data because it avoids overfitting and is highly scalable, so that a typical recommendation problem with millions of input features will have millions of parameters rather than trillionsaws blog
There’s a handy Jupyter notebook you can download to follow along with.
The model equation for factorization machines is defined as:aws blog
Model parameters to be estimated are:
where, n is the input size and k is the size of the latent space. These estimated model parameters are used to extend the model.
If this interests you, head on over to the ML Tutorial and follow along as it creates product recommendations based on the inputs you provide.
You’ll end up with a capability just like the Amazon website, and maybe you’ll become an economic powerhouse in your own right. If you enjoy this machine learning tutorial, feel free to browse the rest of the site for our other high quality AI and ML tutorials and follow-alongs.