Which of the following is$/$are true about collaborative filtering?

User$-$based similarity is usually more stable than item$-$based similarity.

$1$ pts

If you have a dataset of user ratings on some products, you can use these to predict one user's preferences on products that he has not rated.

To use collaborative filtering, you need to manually design a set of features for every item $($e$.$g$.,$ movie$)$ in your dataset, which describes item's most important characteristics

Amazon's item$-$based collaborative filtering requires pre$-$processing.