Products relationship

We are faced with a dataset containing users and the products they use.

Thinking about this a matrix, that is a very sparse one.

Question we want to answer:

• Which are the most frequent combination of products by users?

• How to separate the products in clusters?

---

link