Blog Post

Netflix uses data for a lot more than just recommendations

Stay on Top of Enterprise Technology Trends

Get updates impacting your industry from our GigaOm Research Community
Join the Community!

Netflix is famous for the way it uses algorithms to determine what programs or movies its members might want to watch, but data plays a much broader role inside the company’s streaming service than just informing recommendations. In a blog post on Wednesday, the company explained how it analyzes data to do everything from optimizing playback quality to identifying poorly translated subtitles.

The post, written by Netflix ?director of streaming science and algorithms Nirmal Govind, highlights several areas in which better algorithms could improve the Netflix experience, focusing largely on how to ensure the best-possible playback in any given situation — or, at least, how to ensure users are getting the playback quality they expect. It might be easy enough to find the right theoretical tradeoff between bit rate and rebuffer rates on streaming videos, or to figure out where (geographically) to place which content on the Open Connect content-delivery network, but nothing is that simple in practice.

“[W]e need to determine a mapping function that can quantify and predict how changes in [quality of experience] metrics affect user behavior,” Govind wrote. He continued:

“With vast amounts of data, the mapping function discussed above can be used to further improve the experience for our members at the aggregate level, and even personalize the streaming experience based on what the function might look like based on each member’s ‘QoE preference.’ Personalization can also be based on a member’s network characteristics, device, location, etc.”

However, the most interesting use of data Govind discussed might be how Netflix is using natural-language processing and text analysis to improve the actual quality of the movies and shows it streams. Audio and video quality may be paramount, but the accuracy of closed captions and subtitles is becoming a bigger problem as Netflix expands globally. Some of these issues are identified via Netflix’s own quality checks, but others are peppered throughout scores of member comments and feedback.

The supply chain through which Netflix is trying to optimize quality. Source: Netflix
The supply chain through which Netflix is trying to optimize quality. Source: Netflix

Govind highlights a couple of ways Netflix is trying to solve these problems:

“[W]e can detect viewing patterns such as sharp drop offs in viewing at certain times during the show and add in information from member feedback to identify problematic content. Machine learning models along with natural language processing (NLP) and text mining techniques can be used to build powerful models to both improve the quality of content that goes live and also use the information provided by our members to close the loop on quality and replace content that does not meet the expectations of Netflix members.”

Improving subtitles and captions, and filtering through mountains of comments to find relevant ones, sound like good candidates for the deep learning models that Netflix is experimenting with.

Techniques aside, though, using data to improve the viewing experience is arguably more important to Netflix’s continued success than are accurate recommendations. Yes, the easier it is to find programs you want to watch, the easier it is to watch them. But at the end of the day, Netflix has the same issues as other seemingly invincible companies like Facebook(s FB) and Google(s GOOG) (issues that we’ll delve into in detail at our Structure conference next week): loyalty on the web can be easy come, easy go. If performance starts slipping, those users will start looking elsewhere.

Feature image courtesy of Shutterstock user Twin Design.

6 Responses to “Netflix uses data for a lot more than just recommendations”

  1. Federico Pascual

    Its great to see that Netflix is using natural language processing to improve streamming quality and subtitles. There are much more things that can be done with huge amounts of data and ML/NLP other than recommendation systems. On a side note, if you are interested in Machine Learning, you should check out our new text mining toolkit, we are in private alpha: our tool will be handy to you. Cheers!

  2. If the subtitles are of poor quality, that’s because Netflix hires bad translators who accept being paid shitty money for their work, or mere students that think they speak a language and can translate it. To ensure excellent quality in subtitles, Netflix should pay the translators well. Quality has a price. You pay shit, you get shit.

  3. clark law

    we use netflix, never ever anything new to watch, all old crap that we seen a decade or more , when i,m in the mood for a certain type of movie the selection is rather poor.We have a huge personally DVD collection of 500 or more movies.heavy in horror and scifi, no comedies but a lot of animation about everyone out to date. from disney to dreamworks.This sevice has been a big disapointment, Had Hulu But its glichy and quits playing in the middle of films. Netflix never have this problem.

    • Clark – for $8.99/month are you really complaining? Netflix releases new movies and series all the time. Great documentaries. Not to even mention all the great older series that are great to go back and watch – which is worth the price alone. I have both Netflix and Amazon Prime and find that I can watch most of my content through both.

  4. sionanno

    I work in subtitling and I have to say – it is a very slippery slope to start “improving” subtitles. There are reasons for the choices that are made, and there is a human being, a professional translator, behind most subtitles of any quality. Changes should not be made lightly – it’s easy to think you have a better idea or something that is “closer” to the original, but unless you are a professional who has been specifically hired to do a serious job of bettering a flawed translation, you should refrain from thinking you know better. I’m the first to agree that if there is a mistake in a subtitle, let’s fix it. But we mustn’t let this type of thing get out of hand. Everyone’s a critic. There needs to be more respect for this very creative and skilled profession. I am alarmed to see how little regard there seems to be. For example, our signature lines are being removed from the end of films on Netflix, and in other places too. How dare they? We have worked hard and deserve our credit just like anyone else who had a hand in making the film. And, if there is a query about a subtitle, maybe the person to take it up with would be the translator? I’d like to discuss the issue with someone at Netflix if possible. Sincerely, Sionann O’Neill
    PS – For those who are interested, here’s an article about subtitling that was recently published in The Hollywood Reporter. As it happens, many film directors, after working in close conjunction with their translators, are alarmed and dismayed to learn the translations are being tampered with.