Thursday, December 15, 2005

E=mcCatwomansUx0rd

Not a ton of original stuff this week, sorry -- tail end of that rewrite. Some people have sent me Harold Pinter's speech and asked if I had a comment. To which I reply: "Thanks, but he is HAROLD GODDAM PINTER. He is so mind-numbingly beyond my meager abilities that all I can doo is 'ook' at him like the monkey in front of the obelisk in 2001. And not even the smart monkey. The other one."

Via Warren, we find that an Oklahoma State University neural network has been trained to recognize what makes a successful movie.

Using data on 834 movies released between 1998 and 2002, Sharda found that the neural network can judge a film based on seven key parameters: the “star value” of the cast, the movie’s age rating, the time of release against that of competitive movies, the film’s genre, the degree of special effects used, whether it is a sequel or not, and the number of screens it is expected to open in. This allowed it to place a movie in one of nine categories, ranging from “flop” (total takings less than $1 million) to “blockbuster” (over $200 million).


Catwoman was put into the matrix. The equations produced are excerpted in the title. Dammit, this is what I get for turning my back on science and plunging into screenwriting.

12 comments:

Anonymous said...

Damn straight. Pyshics major and he starts writing movies. What the hell were you thinking?

Anonymous said...

*physics

(I can spell, don't hate me)

:)

david golbitz said...

Speaking of Catwoman, did DC have to use that movie in the solicitation for the upcoming Blue Beetle comic?

BLUE BEETLE #1
Written by Keith Giffen & John Rogers
Art and cover by Cully Hamner
Keith Giffen (FORMERLY KNOWN AS THE JUSTICE LEAGUE) and hot screenwriter John Rogers (Catwoman) team with Cully Hamner (RED, BATMAN: TENSES) to bring back the Beetle!


I thought the idea was to get people excited for the comic.

(And yes, I'm aware that what John wrote and what ended up on the screen are two vastly different things. I kid. I kid because I love.)

Karl said...

Article says he's expanding it later to include DVD sales...
That should have been all ready included factor.

Unknown said...

That should have been all ready included factor.

True, but those numbers are almost impossibly guarded. Now that he has a studio in, he'll have a way in to get those figures.

Anonymous said...

The brain might be more useful if it could predict profit margins or returns as multiples of expenses. A 200 million dollar movie that sells 200 million dollars in tickets isn't really a blockbuster, though I expect most do OK once foreign and DVD money arrive.

GM Doug said...

Hey Shan, I'm a physics major (well we don't have "majors" over in the UK we arrive on Day 1 - having picked our college course a year before - and just do Physics and all that entails... maths, electronics, yada yada yada.) and believe me there comes a point in about 3rd year that you realise you REALLY want to be doing something else.

Anything else.

Hell I considered teaching and of all things becoming an actuary. And I hated maths.

So rogers made the leap all of us Physics geeks so wanted to do.

Because really... while we can imagine far out stuff in our heads... it's far far cooler to see it appear on screen with physics we know it's going to work - but damn it Lightsabres, Lasers, flying aliens from Krypton and Magic are just too damn fun.

As for me... I ended up in an office writing VB apps for my company and in the spare time I have run a Football team.

But I always wanted to be a movie writer...

"Leaping from tree to tree as they float down the mighty rivers of British Columbia! The Fir! The Larch! The Redwood! The mighty Scots Pine!"

Back to the point in this... Has this guide at any point predicted results over the course of a full year? Or is just matching data already out there?

Anonymous said...

2001... computer that can pick hit movies...

It's so obvious where this is going: A mad computer hijacks Hollywood and tricks studios into making movies only a machine could love.

Since this holds special interest for blockbuster-picking automatons, I'm sure it will score high.

Dweeze said...

I took the expanding it to include DVD sales to mean that he is expanding the model to predict DVD sales, not to take DVD sales into account when predicting box office success. After all, the vast majority of DVD sales don't take place until box office success or failure has already been determined.

Anonymous said...

that girl: You'll notice that it was trained on data from just 1998-2002, presumably because the relative importance of these factors in success have changed significantly over time with shifts in the movie market. So, data from 1998-2002 was used as one can assume that period is both relatively static and representative of the current state.

I'd expect that, using the model derived from the 1998-2002 data, its output for anything before 1980 would be garbage, even after adjusting for inflation.

Karl said...

Oy, my grammar...

I didn't realize the dvd sales were that guarded.

Unknown said...

The principal thing you need to do is to settle on a cognizant choice to escape this circumstance. Choose to change and turn into a butterfly as opposed to a fly got up to speed on the payday advance web. check cashing near me open now san diego