This Psychologist Might Outsmart the Math Brains Competing for the Netflix Prize • Hacker News Board

Re: This Psychologist Might Outsmart the Math Brains Competing for the Netflix Prize

#11

Post by elq » Thu, Feb 28, 2008, 6:29 PM UTC

tel wrote: post #4

Now there just needs to be an economist to drop in and sneak an entry with a 9% improvement. Statistics and math are excellent tools, but they're quite blind.

well... there is probably already a few teams with econometricians on board. based on my prior experience with that ilk - I'm sure their progress got held up by the insistence that all returns (err, ratings) follow a guassian distribution.

Also, all of that training in math and stats is pretty useless if you just use it to "prove" something you already "know" :)

Re: This Psychologist Might Outsmart the Math Brains Competing for the Netflix Prize

#12

Post by jwp » Thu, Feb 28, 2008, 6:45 PM UTC

hugh wrote: post #5

I've always known that science journalists don't know anything about science. I'm only now realizing that they don't know anything about journalism either. They only know how to tell one story. That's the story of the underdog scientist who is (or more commonly, "might be about to") upsetting the stuffy old establishment. Everything that goes on in the scientific community has to be shoehorned into the storyline of "…

Sure, there's plenty of fluff in the article, but it isn't because the author (http://www.math.wisc.edu/~ellenber/) is clueless about science. The article is probably at the right level of technical depth. Seems ambitious to even touch SVD in Wired.

Re: This Psychologist Might Outsmart the Math Brains Competing for the Netflix Prize

#13

Post by gms » Thu, Feb 28, 2008, 7:14 PM UTC

hugh wrote: post #5

I've always known that science journalists don't know anything about science. I'm only now realizing that they don't know anything about journalism either. They only know how to tell one story. That's the story of the underdog scientist who is (or more commonly, "might be about to") upsetting the stuffy old establishment. Everything that goes on in the scientific community has to be shoehorned into the storyline of "…

I'm willing to bet every dollar to my name that the author can eat you for breakfast when it comes to "knowing about science".

Re: This Psychologist Might Outsmart the Math Brains Competing for the Netflix Prize

#14

Post by brent » Thu, Feb 28, 2008, 7:22 PM UTC

Why does he or the author consider him a psychologist? I think its a bit misleading when simply reading the title of the article. He has an undergraduate degree in psychology, but his master's is in operations research (and correct me if I'm wrong, but OR is usually the application of statistics to business problems, right?). Wouldn't the article more accurately be something like "Operations research guy uses statistical technique motivated by psychology ... "

Re: This Psychologist Might Outsmart the Math Brains Competing for the Netflix Prize

#15

Post by hugh » Thu, Feb 28, 2008, 8:16 PM UTC

jwp wrote: post #12

hugh wrote: post #5

I've always known that science journalists don't know anything about science. I'm only now realizing that they don't know anything about journalism either. They only know how to tell one story. That's the story of the underdog scientist who is (or more commonly, "might be about to") upsetting the stuffy old establishment. Everything that goes on in the scientific community has to be shoehorned into the storyline of "…

Sure, there's plenty of fluff in the article, but it isn't because the author ( http://www.math.wisc.edu/~ellenber/ ) is clueless about science. The article is probably at the right level of technical depth. Seems ambitious to even touch SVD in Wired.

Oops. I stand corrected. The author of this particular article obviously does know the subject very well. My apologies.

Re: This Psychologist Might Outsmart the Math Brains Competing for the Netflix Prize

#16

Post by technoguyrob » Thu, Feb 28, 2008, 8:31 PM UTC

gms wrote: post #13

hugh wrote: post #5

I've always known that science journalists don't know anything about science. I'm only now realizing that they don't know anything about journalism either. They only know how to tell one story. That's the story of the underdog scientist who is (or more commonly, "might be about to") upsetting the stuffy old establishment. Everything that goes on in the scientific community has to be shoehorned into the storyline of "…

I'm willing to bet every dollar to my name that the author can eat you for breakfast when it comes to "knowing about science".

Umm...the author is a math professor at the University of Wisconsin, and he's a published author (in fiction).

Jordan Ellenberg (ellenbergwired@gmail.com) is a math professor at the University of Wisconsin and author of the novel The Grasshopper King.

Re: This Psychologist Might Outsmart the Math Brains Competing for the Netflix Prize

#17

Post by gms » Thu, Feb 28, 2008, 8:49 PM UTC

technoguyrob wrote: post #16

gms wrote: post #13

Earlier quoted context omitted.

I'm willing to bet every dollar to my name that the author can eat you for breakfast when it comes to "knowing about science".

Umm...the author is a math professor at the University of Wisconsin, and he's a published author (in fiction). Jordan Ellenberg (ellenbergwired@gmail.com) is a math professor at the University of Wisconsin and author of the novel The Grasshopper King.

...yes...that's what I was pointing out...

Re: This Psychologist Might Outsmart the Math Brains Competing for the Netflix Prize

#18

Post by andreyf » Thu, Feb 28, 2008, 9:07 PM UTC

Yet another psychological problem mathematicians too eagerly claimed as their own. Why not try to find patterns about the movies which appeal to a certain individual? Is it against the rules to use outside data (actors, directors, etc)?

Re: This Psychologist Might Outsmart the Math Brains Competing for the Netflix Prize

#19

Post by hugh » Thu, Feb 28, 2008, 9:20 PM UTC

hugh wrote: post #5

I've always known that science journalists don't know anything about science. I'm only now realizing that they don't know anything about journalism either. They only know how to tell one story. That's the story of the underdog scientist who is (or more commonly, "might be about to") upsetting the stuffy old establishment. Everything that goes on in the scientific community has to be shoehorned into the storyline of "…

Whoops! OK, so just to clarify, I didn't mean for my off-hand generalization about "science journalists not knowing anything about science" to extend to the author of this article in particular, who as several people have now pointed out, turns out to actually be a professor of mathematics at Madison.

Re: This Psychologist Might Outsmart the Math Brains Competing for the Netflix Prize

#20

Post by andreyf » Thu, Feb 28, 2008, 9:23 PM UTC

andreyf wrote: post #18

Yet another psychological problem mathematicians too eagerly claimed as their own. Why not try to find patterns about the movies which appeal to a certain individual? Is it against the rules to use outside data (actors, directors, etc)?

From the Netflix prize FAQ:

Why not provide other data about the movies, like genres, directors, or actors?

We know others do. Again, Cinematch doesn’t currently use any of this data. Use it if you want.

That seems like an easy target - if I've 5-starred every movie with Kevin Spacey, I probably will like anything with Kevin Spacey. Why not mine blogs and reviews, trying to find themes which are appealing to individuals?