Reverse-engineering Hollywood

How to bypass Netflix and create your own custom search genres
January 6, 2014 by Amara D. Angelica

One result from a made-up “altgenre”: Cult Evil Kid Horror Movies (credit: Netflix)

If you’re a combo Netflix/Internet-text-algorithm-obsessed geek like me, you will totally love this amazing article in The Atlantic.

Turns out Netflix has created a database of 76,897 micro-genres that offer a peek into the American psyche, The Atlantic senior editor Alexis Madrigal has discovered, using a program called UBot Studio to scrape every single one of them and then deconstruct the system.

Hidden syntax revealed

“As the thousands of genres flicked by on my little netbook, I began to see other patterns in the data,” he said. “Netflix had a defined vocabulary. The same adjectives appeared over and over. Countries of origin also showed up, as did a larger-than-expected number of noun descriptions like Westerns and Slashers. …”

Here’s a snippet of Madrigal’s snippet of altgenres he saw:

Mind-bending Cult Horror Movies from the 1980s
Dark Suspenseful Sci-Fi Horror Movies
Gritty Suspenseful Revenge Westerns
Violent Suspenseful Action & Adventure from the 1980s
Time Travel Movies starring William Hartnell

He went even further, working with Georgia Tech professor and Atlantic contributing editor Ian Bogost to build The Atlantic‘s Netflix-Genre Generator, capable of creating such imaginary “altgenres” as “Travel Sci-Fi Movies Based on Bestsellers About Friendship” — to mention one I just created from the three keywords used in their genre-generator demo.

Ultimate movie producer’s research tool

Madrigal also found that his system could even identify altgenres for movies that didn’t exist, but should. As has Netflix, which now has a “terrific advantage in their efforts to produce their own content: Netflix has created a database of American cinematic predilections. The data can’t tell them how to make a TV show, but it can tell them what they should be making. When they create a show like House of Cards, they aren’t guessing at what people want.”

Using large teams of people specially trained to watch movies, “Netflix has meticulously analyzed and tagged every movie and TV show imaginable,” Madrigal’s delightfully OCD research has uncovered. “They possess a stockpile of data about Hollywood entertainment that is absolutely unprecedented.”

But most intriguing, he said, were the subjects — “a complete list of which form a window unto the American soul:”

(Credit: Alexis Madrigal/The Atlantic)

“Netflix has built a system that really only has one analog in the tech world: Facebook’s NewsFeed,” he said. “But instead of serving you up the pieces of web content that the algorithm thinks you’ll like, Netflix is serving you up filmed entertainment. Which makes its hybrid human and machine intelligence approach that much more impressive.”

Best part: once you know the syntax he describes, you can create your own altgenres and search Netflix for them, instead of being limited to genres that Netflix has assigned to you, based on what you watch. Oh yeah.

Amara D. Angelica is Editor of KurzweilAI