Adam Fields (weblog)

This blog is largely deprecated, but is being preserved here for historical interest. Check out my index page at adamfields.com for more up to date info. My main trade is technology strategy, process/project management, and performance optimization consulting, with a focus on enterprise and open source CMS and related technologies. More information. I write periodic long pieces here, shorter stuff goes on twitter or app.net.

1/9/2005

Numbered recommendation systems are missing some semantic meanings

Filed under: — adam @ 8:08 pm

Netflix asks me to rate a movie on a scale of 1-5. Where’s “I really wanted to like this movie, but it just dragged on and on” (Hero) or “I wasn’t in the right mood, but I could have liked it” (Hero), or “blah blah blah really pretty, but where’s the plot” (Hero), or even “it really sucked, but at least we got a good laugh – had I watched it alone, I might have clawed my eyes out” (King Arthur).

These indicate that I might like other movies in the same karass, but I feel bad giving them a high rating, because I didn’t actually like them.


One Response to “Numbered recommendation systems are missing some semantic meanings”

  1. Richard Terry Says:

    This site (http://recommendz.mcgill.ca:8080/prototype/) does something similar to what you talk about. They allow flexibility in how you rate a movie. (My favorite is “sword fighting goodness”.) When I switched from Netflix to Walmart’s DVD service, I found that I missed Netflix’s excellent recommendations. I currently us this site: http://movielens.umn.edu/. I also tried this one (http://www.whattorent.com/), but it didn’t suit me.

Powered by WordPress