Item-item CF

User-based CF works quite well and is widely used in production in the wild, but it does have a few considerable downsides. First, it's difficult to get enough preference information from users, leaving many of them without a solid base for relevant recommendations. Second, as the platform and the underlying business grows, the number of users will grow much faster than the number of items. Netflix, for example, to keep the discussion in the familiar area of movies, grows its user base massively by expanding into new countries, while the production of movies stays pretty much the same on a yearly basis. Finally, the user's data does change quite a lot, so the rating matrix would have to be updated often, which is a resource-intensive ...

Get Julia 1.0 Programming Complete Reference Guide now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.