Netflix—beautiful, beautiful Netlix—is, experts hypothesize, the pinnacle of the human experience. It doesn't get much realer than finishing your marathon of House of Cards only to have that incredible Movie Suggestion Machine take your hand and lead you through three seasons of Luther.
But, for as wonderful as it is, the complicated magic behind Netflix's taste-predicting algorithms has always been shrouded in secrets. No more, though, because Wired managed to track down the guys behind the science/math/Fancy Guessing Robot and get them to explain exactly what the hell is going on.
It turns out that Netflix employs more than 40 analysts that tag movies and television shows to help create patterns based on similarities. Director, theme, star, year produced, whether the cast and crew prefered turkey or ham for on-set lunches. It's all taken into consideration. They also take your user habits—how you navigate through the site and scroll pages and genres—into account, infering that people with similar tendencies will have an overlap in taste.
Does Netflix keep track of my viewing?
Amatriain: We know what you played, searched for, or rated, as well as the time, date, and device. We even track user interactions such as browsing or scrolling behavior. All that data is fed into several algorithms, each optimized for a different purpose. In a broad sense, most of our algorithms are based on the assumption that similar viewing patterns represent similar user tastes. We can use the behavior of similar users to infer your preferences.
So if I’m viewing on my iPad at midnight, do I see different recommendations than I would on my TV at 8 pm?
Amatriain: We have been working for some time on introducing context into recommendations. We have data that suggests there is different viewing behavior depending on the day of the week, the time of day, the device, and sometimes even the location. But implementing contextual recommendations has practical challenges that we are currently working on. We hope to be using it in the near future.
The result is that your high-tech moving pictures box knows what you want to watch before you know you want to watch it.
Thanks for being you, Netflix.
P.S. Do yourself a favor and watch the most recent Netflix original series, Orange Is the New Black. [Wired]