[ Digital cinema, democratization, and other trends remaking the movies ]

AD: Fans, Friends & Followers

Thursday, March 27, 2008

Movie Recommendation Technology

A piece I wrote last year for the site FilmInFocus is now available; it deals with using software to try to predict what movies we'll enjoy.

From the piece:

    One problem that the technology hasn't yet begun to deal with is our changing moods. "You might enjoy one movie when you're alone and feeling down, but things are different when you're going to be with a bunch of people," [University of Minnesota professor Joe] Konstan says. There's no way yet to tell Blockbuster that you're feeling burnt out on a Friday night and need some light entertainment, or that you owe your girlfriend a "date night" movie, or that you're in a documentary frame-of-mind.

    Another issue is that none of the sites you visit share information about your tastes with anyone else. "Blockbuster thinks about the ratings you've given it as their information," explains Toffer Winslow, an executive at ChoiceStream, the software company that supplies Blockbuster with recommendation technology. He says that users may feel like their privacy has been compromised if information they've given to one site is used by another to serve up more personalized content.

Among the companies I mention are Netflix, ChoiceStream, Matchmine, and Amazon.

Labels: , , , , ,


  • A couple of years ago I had a sudden desire to see some of the old Kung Fu films from the '70s. I bought a couple from Amazon. Now, NO MATTER WHAT I DO, Amazon keeps recommending Kung Fu films to me, even thought I would be perfectly happy if I never saw another one. The recommendation software can sometimes seem miraculous but is, in my estimation, just as likely to be moronic.

    By Blogger Unknown, at 2:09 AM  

  • My fiancĂ© and I sharing a Netflix account makes the whole thing that much more hilarious. We're constantly getting things like:

    Because you enjoyed


    we think you'll enjoy


    A bit of an exaggeration, but not that far from the truth. If only there were a way for us to agree on ratings. "Did you rate Flight of the Conchords?" "I don't remember, why?" "It's five stars." "What would you rate it?" "...probably five stars." "Well, then it doesn't matter, does it?" "...I think it was you."

    Whatever helps you sleep at night, I guess.

    By Blogger Liz, at 1:52 AM  

  • "recommendation" techniques have been criticized for exactly the problem mentioned in these comments ... and for the reasons you highlight in your post, Scott. I will get and send you the essays from two economics professors at Wharton that show that "recommendation" techniques as implemented actually tned to narrow the range of choices of a general population, even though an individual member of the group may feel like he has been pointed in a new direction.

    At IndiePix, we have implemented a way of browsing our catalog of over 3100 titles based on our Discovery Engine. The process is one of exploring and discovery -- which explicitly incorporates user mood, interest, phase of moon, etc! Because it is the user who chooses; we don't recommend; we facilitate exploring.

    Try it at ... or if you want to see a specific example, try the Seventh Seal at

    Of course, there may always be times, when you don't even want to choose!

    By Blogger bob in ny, at 5:27 AM  

Post a Comment

<< Home