This was the basic idea, and with the advent of big data, the CF technique has become quite powerful. As it's been applied to different business needs and usage scenarios, the algorithm was refined to better address the problems it was attempting to solve. As a consequence, a few other approaches emerged, and the original one became known as user-item CF.
It's gotten this name because it takes as its input user data (user preferences, rankings) and outputs item data (item recommendations). It's also known as user-based CF.
You can see it illustrated in the following diagram:
The preceding diagram shows that Annie ...
