Netflix Recommendations Are More Complicated Than You Might Think


Although it may seem as though Netflix displays its recommendations at random, the company insists there’s a very complex system quietly at work behind-the-scenes.

Anyone who has received suggestions such as The Toxic Avenger or I Spit on Your Grave after watching James Cameron’s Titanic may think Netflix is just being stupid. However, there’s actually a very complicated algorithm that determines which movies you might be interested in checking out.

During a recent chat with Wired magazine, two Netflix engineers revealed that the company is watching your every movem during your visit to the streaming service. Everything from how you scroll through the content to when you watch movies is used to make recommendations.

“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,” engineering director Xavier Amatriain explained.

He continued, “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.”

In other words, Netflix knows that you’re the sort of person who might watch The Toxic Avenger even if you’ve never really considered the idea. You might not be the biggest horror nerd on the planet, but Netflix knows you’ve spent some time browsing through the scary movies.

Carlos Gomez-Uribe, vice president of product innovation and personalization algorithms, admitted that your ratings really don’t matter much in the long run. At the end of the day, what you watch matters more than the two-star review you gave to View from the Top.

“Testing has shown that the predicted ratings aren’t actually super-useful, while what you’re actually playing is,” Gomez-Uribe explained.

Earlier this year, Netflix rolled out Max on the Playstation 3. Instead of pouring over all of the stuff the company has to offer on its streaming service, this “mood ring”-style program finds something for you to watch after asking a series of calculated questions.

“Sometimes, when Max feels particularly confident, he will offer a fun, personalized suggestion after asking only one simple question. Other times, when intoxicated by past success, Max will get a little cocky and offer a suggestion right away, no questions required,” the company explained.

What do you think about the Netflix recommendation system? Are you surprised that the company tracks every move you make on the service to figure out what you might enjoy?

[Image via Shutterstock.com]

Share this article: Netflix Recommendations Are More Complicated Than You Might Think
More from Inquisitr