5 ms·
Just so you know, 1.recommender systems != collaborative filtering. 2.RBMs are good only for certain situations like explicit ratings (where users provide dire
by harshgupta 8y ago
Just so you know,
1.recommender systems != collaborative filtering.
2.RBMs are good only for certain situations like explicit ratings (where users provide direct feedback), its a really hard problem in implicit recommendations (which are most sites). See Bayesian Personalised Ranking Algorithm.
3. >I'd argue that the logistics of shipping perishable food is significantly more difficult than streaming an audio file. Thats exactly the point. I am not measuring hardness. But that looks a supply chain problem that a software engineering problem.