Dragon Ball Z Uncensored

MADali · (Mon Apr 28, 2008 9:36 am)

Let's say you have 10 movies which you have rated 1-5. And you have 5 other friends with similar movies rated their way. How can you see how similar the relationship is?

Like if Person A and Person B gave Oldboy 5, then the relationship on that movie would be 100%. But what if one person gives it 4 and other 3? Or one 1 and other 4? It seemed so simple in my mind, but I cant figure how to calculate that

Magic Juan · (Mon Apr 28, 2008 9:52 am)

So you're trying to figure out how to analyze the intensity of a friendship by their taste in movies?

Magic Juan · (Mon Apr 28, 2008 9:53 am)

Didn't we do this already?

FancyMichael · (Mon Apr 28, 2008 10:10 am)

I can't decide whether I should go pick up GTA4 or not.

It's getting such wonderful reviews!

Servbot · (Mon Apr 28, 2008 10:34 am)

You're such a comformist.

Theldorrin · (Mon Apr 28, 2008 10:34 am)

It's pretty simple.

For every two people, you take the difference in each of their ratings and sum up that difference. The larger the number the more dissimilar.

That's pretty much the only thing you can do with just basic numbers, although I suppose you could always normalize rankings to a curve. So that if someone gives a movie a 4, but it's their highest ranked movie, then it would be treated as a 5. That's not necessarily a wise course of action on a limited set, though.

YES · (Mon Apr 28, 2008 10:41 am)

I guess people tend to like the movies I like, but I never like the movies everyone else likes, so that means I like fewer movies than everyone else

Anyway, fuck thsi thread

cloacal kiss · (Mon Apr 28, 2008 11:25 am)

I think 10 movies rated on a scale of 1-5 would provide very little information about people and their potential interaction.

MADali · (Mon Apr 28, 2008 11:29 am)

Theldorrin, that seems like a simple solution, and I guess the easiest way to go.

I'll get back to you with more questions, I guess. I'm thinking of starting a sort of new website with a friend.

Theldorrin · (Mon Apr 28, 2008 11:29 am)

I disagree. I think it's quite possible that it could contain a lot of information if the right movies are picked.

cloacal kiss · (Mon Apr 28, 2008 11:45 am)

Shut up bitch

Theldorrin · (Mon Apr 28, 2008 11:56 am)

More complicated solutions really need more data, MADali.

Psarevo · (Mon Apr 28, 2008 12:16 pm) Reply

The last thing Dada needs is more complication!

Psarevo · (Mon Apr 28, 2008 12:17 pm) Reply

Oh I'm sorry, you said data.

Psarevo · (Mon Apr 28, 2008 12:18 pm) Reply

*killfiles self*

Ryoko's Biatch · (Mon Apr 28, 2008 12:29 pm)

Hey, MADali, why don't you just use that gay fucking application on Facebook?

MADali · (Mon Apr 28, 2008 12:47 pm)

I'll tell you exactly what I plan to do, Theldorrin, later I guess. I need to sort out my plans.

Theldorrin · (Mon Apr 28, 2008 12:48 pm)

I think you might be able to use some of those Netflix algorithms if you have tons of data. But that's more predictive oriented.

L Ron Butterfly · (Mon Apr 28, 2008 2:28 pm)

This thread is about al gore rhythms.

Big Fagot · (Mon Apr 28, 2008 3:22 pm)

A particular algorithm of this nature would be anywhere between some basic piece of shit and a million dollar trade secret.