Glossary · Phase 3 · Machine learning
Matrix factorization
Learning a short vector for every user and item so that their dot product predicts the user's rating of the item.
Try it now
The same ratings, one row per rating. Group them by item, then switch between the mean and the count.
Learn it properly
Recommender systems · Shrunk averages, item-to-item collaborative filtering, matrix factorization, and cold start. About 35 minutes.
Other terms from the same lesson
- cold start: A new item or user with no behavior yet, which collaborative filtering can't place; content features and exploration fill the gap.
- collaborative filtering: Recommending from behavior alone: items are similar when the same people rated or bought them alike.
Never trained an AI?
Teach a small one to write, right in your browser. It takes about 20 seconds and needs no account.