advertisement
advertisement
advertisement
  • 11.05.13

The Video Recommendation Engine Behind The Xbox Launches A Standalone iPad App

Jinni uses what it calls the “Entertainment Genome” to understand individuals’ movie and TV tastes.

Two months after inking a deal to power Xbox’s recommendation engine, Tel Aviv startup Jinni has launched a standalone iPad app and revamped website to help users discover movies and shows across their TV and subscription services.

advertisement

Much like how Pandora is based on the Music Genome Project, Jinni uses what it calls the Entertainment Genome, a project that maps mood, style, plot, setting, and other “genes” to understand individuals’ movie and TV tastes. Its engine is also behind Comcast and Vudu’s discovery tools.

Take, for instance, the movie Life of Pi. IMDb might categorize this as a movie that crosses the adventure, drama, and fantasy genres, but Jinni affixes much more specific tags: It’s a coming-of-age story of a young hero’s survival against the odds of living at sea. The movie explores such themes as confinement, hope, and danger, and the mood is stylized, captivating, contemplative, and touching. “Any other engine or guide, they get the basic metadata that comes with content–that it’s an adventure or drama, or based on a book maybe. You have very limited information,” media and marketing manager Nikki Ralston told Fast Company. “Here, Jinni really understands each content item like a movie expert would.”

Jinni can import the ratings from other websites, including Netflix, Facebook, Flixter, IMDb, and Rotten Tomatoes, analyzing them to find the entertainment genes that appeal to users. The website and iPad app also surfaces recommendations specific to users’ subscription services and television providers.

Furthermore, the service can leverage users’ social connections to find “taste buddies,” or friends with similar viewing preferences. “Using the general social network is a really, really bad way to do recommendations, but we can pick out those couple of people in your social network that have really similar tastes,” Ralston said. Looking across different profiles, Jinni can also analyze the cross-section of people’s tastes to recommend movies for groups.

“Now when you come home from work, instead of flipping through your favorite 15 channels you happen to like at the moment … you can get exactly what you want to watch within a few clicks,” she said.

About the author

Based in San Francisco, Alice Truong is Fast Company's West Coast correspondent. She previously reported in Chicago, Washington D.C., New York and most recently Hong Kong, where she (left her heart and) worked as a reporter for the Wall Street Journal.

More