By John D. Sutter, CNN
(CNN) - Netflix recently lifted the hood on its recommendation algorithm, which helps people who use that video-streaming service to find movies and shows that they may not know about but that they would like.
The most interesting stat the company listed on its corporate blog: 75% of the videos people watch are found via some kind of recommendation that the company employs on its apps and website. That means most pieces of content people end up watching on Netflix are found with the help of a computer equation that thinks it know what you want to watch - or via Facebook, for people outside the United States (Netflix doesn't connect with Facebook in the U.S. because some say the Video Privacy Protection Act prohibits video rental records from being shared. There is some confusion over when and how that law applies.)
In any event, here's what Netflix has to say about the popularity of its digital recommendations:
We have adapted our personalization algorithms to this new scenario in such a way that now 75% of what people watch is from some sort of recommendation. We reached this point by continuously optimizing the member experience and have measured significant gains in member satisfaction whenever we improved the personalization for our members.
That's pretty remarkable. But the company admits it still has some kinks to work out.
One of them: Tailoring recommendations to individual Netflix users instead of to the entire household. Often, a family will sign up for a single Netflix account, so recommendations sometimes get muddled between what mom, dad and the kids want to watch. Netflix says in its Friday blog post that it addresses this issue by showing a "diversity" of content that would be appropriate for any member of the family.
It is important to keep in mind that Netflix’ personalization is intended to handle a household that is likely to have different people with different tastes. That is why when you see your Top 10, you are likely to discover items for dad, mom, the kids, or the whole family. Even for a single person household we want to appeal to your range of interests and moods. To achieve this, in many parts of our system we are not only optimizing for accuracy, but also for diversity.
Do you use Netflix's recommendation engine - or others like it on sites like Amazon? If so, what do you think? Where do these equations fall short and when have they gotten it right?
Each week, CNN's "The Next List" profiles innovators, visionaries and agents of change. They’re not household names just yet, but they’re movers and shakers in their own worlds. We’re introducing them to you because these individuals are steadily mapping the course to the future with their new ideas.
WHO: Host Dr. Sanjay Gupta
WHEN: Saturdays at 2:30 p.m. ET - All new time!
WHERE: Only on CNN
© 2012 Cable News Network. Turner Broadcasting System, Inc. All Rights Reserved.
| Advertising practices | Advertise with us | About us | Work for us | Help
Thanks . I have been looking for something like this . Wonderful information I will check back for more info about juicing and recipes.
My brother recommended I would possibly like this blog. He was once totally right. This post truly made my day. You can not imagine just how a lot time I had spent for this information! Thank you!
What? NOOOOOOOOOOOOO! (That was a scream.) Not only did I like that store, I also am hoiferrid I did not know about a chance to get cheap DVDs. I posted Video Hut praise about a year ago (when I was still organized enough and disciplined enough to be succinct): ": Like the Early 90's" I hope it's not bad form to link to one's blog in a comment. Anyway, SOS, thanks for the heads up up Video Hut, even tho it's bad news. I have to admit, I have not been to a video store since we got Apple TV. Pretty bad selection, but I'm waiting for Xbox/Netflix rental thing to start.
Excellent items from you, man. I have take into accout your stuff previous to and you are just extremely magnificent. I actually like what you've bought here, certainly like what you are stating and the best way by which you say it. You're making it enjoyable and you still take care of to stay it sensible. I can't wait to learn far more from you. This is actually a terrific web site.
I use Netflix but do not like their "recommendation filter". Either newly acquired movies don't show up or it's full of things I would never watch. Seems like they are trying to sell us things they have been told to "push".
I really like the beer recommendations at http://www.Pintley.com. ... Way better than the movie recommendations on Netflix in my opinion.
Get off ur duffs Netflix!!!! make that distribution with HBO considering that Starz pulled over 900 movies from Netflix library in February!!! can only watch C.H.U.D. but sooo much......
HOW YOU GOING TO RECOMMEND SOMETHING THAT YOU DO NOT HAVE. WHAT I RECOMMEND IS THAT YOU BREAK BREAD WITH THE FILM INDUSTRY, PAY A LITTLE BIT EXTRA FOR THE MOVIES AND DELIVER! IT MAKES NO SENSE, DID WHEN YOU HAD THE CROWN, BUT YOU'VE BEEN STRIPPED AND DO NOT KNOW HOW TO HANDLE IT, MAN UP AND GET IT ON!
love their recommendations. It's one of the main reasons I use them and not amazon (even though i am a prime customer)
I want to be able to browse outside of what is recommended and I find Netflix's Browse, Browse by Genre, Browse by ??? and Search capabilities very lacking.
I use the recommendations all the time. They're great at finding new movies and shows that you haven't heard of before.
DR, your math may work out well, but that still doesn't get at the human aspect of eojyning a film (or detesting it, for that matter). What is at issue here is Netflix's claim that they can accurately quantify the quality of a movie as it compares to customers' tastes. That's a noble and bold claim. Suffice to say I have stumbled across some really interesting plot descriptions on the Netflix recommendations, but have yet to say that it's pointed out a hidden gem to me. I can usually tell which elements of the recommended film brought it up on my profile's radar, but there's more to a good movie than a algorithmic string of similar elements. I guess I approach the recommendations with a healthy dose of Caveat Emptor. I know when I am watching a recommendation from Netflix that I may send it back without finishing it to hurry along that next movie in my queue. Kieran
Yeah, and Hollywood movies by and large suck eggs anyway.