22 November 2008

Machine learning in the news

The New York Times published a feature about the Netflix challenge and some of the people who are racking their brains working on it. One major problem to surmount is, apparently, the film "Napoleon Dynamite":

The reason, Bertoni says, is that "Napoleon Dynamite" is very weird and very polarizing. [...] It’s the type of quirky entertainment that tends to be either loved or despised. The movie has been rated more than two million times in the Netflix database, and the ratings are disproportionately one or five stars. [...]

Amazingly, Bertoni has deduced that this single movie is causing 15 percent of his remaining error rate [...] if Bertoni could anticipate whether you’d like "Napoleon Dynamite" as accurately as he can for other movies, this feat alone would bring him 15 percent of the way to winning the $1 million prize.

The NYT also gets geek points for essentially accurately explaining singular value decomposition in layman's terms on page 4 of the article.

In other news, researchers are trying to save humans from the drudgery of watching and annotating soccer games with hidden Markov models:

Detection of goal event in soccer videos [PDF], Hyoung-Gook Kim et al., Department of Communication Systems, Technical University of Berlin.

No comments:

Post a Comment