Netflix Research: Machine Learning Platform




[ad_1]

We invest heavily in machine learning to continually improve our member experience and optimize the Netflix service end-to-end. As researchers, we innovate using machine learning in many areas where we research, design, implement, evaluate, and productionalize models and algorithms through both offline experiments and online A/B testing.

Source


[ad_2]

Comment List

  • WeAreNetflix
    December 16, 2020

    Ok but how does your algorithm work?

  • WeAreNetflix
    December 16, 2020

    Not me 😁

  • WeAreNetflix
    December 16, 2020

    What are your thoughts about people using numeric codes to explore categories more thoroughly. Doing this (as you probably know) sometimes brings up content that is never shown using the conventional search methods, which is why I Guess people do this (as well as to refine sub categories/genres more effectively for example; 'Werewolf Movies' or 'Haunted House, Movies'. I've tested it myself and wonder why a specific for example; (say Horror Movie because I am a Horror Movie Fan) Horror movie shows up that has not been presented to me in the conventional search methods. You would think that they would be somewhere in the search methods provided by Netflix. Has this have anything to do with the search algorithms programmed into your technology? I notice that when I get hooked on a certain subgenre (again lets just say 'Haunted House Movies') Recommendations for me will be maybe 50 or 60 percent when really it should be in the 80 or 90 percent level and some are never recommended to me at all. I just find them by accident. I Typed in 'Classic Horror Films' and it brings up Films such as 'Blazing Saddles' (Ok so maybe you can laugh yourself to death), 'That 70,s Show' and 'Hot Fuzz' but to name a few. These are my questions and my frustrations.

Write a comment