Matrix factorization model

User 2570 | 12/23/2015, 5:54:42 PM

Matrix factorization has found incredible success and widespread application as a collaborative filtering ? but on the documentation does not show how to deal with new users, new items(cold start problem), and new interactions (e.g. ratings, clicks, etc.).

Comments

User 2646 | 12/23/2015, 6:59:11 PM

@erigits our documentation addresses the three items you mentioned here:

https://dato.com/learn/userguide/recommender/making-recommendations.html#making-recommendations-for-new-users

https://dato.com/learn/userguide/recommender/making-recommendations.html#incorporating-new-observation-data