Developing a Vandalism Detector For Wikipedia

marpot writes "In an effort to assist Wikipedia's editors in their struggle to keep articles clean, we are conducting a public lab on vandalism detection. The goal is the development of a practical vandalism detector that is capable of telling apart ill-intentioned edits from well-intentioned edits. Such a tool, which will work somewhat like a spam detector, will release the crowd's workforce currently occupied with manual and semi-automatic edit filtering. The performance of submitted detectors will be evaluated based on a large collection of human-annotated edits, which has been crowdsourced using Amazon's Mechanical Turk. Everyone is welcome to participate."

Comments

inb4

[Anonymous Coward]

it's basically a regex for:

Children are lovely before they get ugly, and learn to do bad stuff.
Flowers are pretty before they get shitty, and rot and turn to dust
I heard that life is a wheel and you can't make it stop
If you try it'll flatten your head
It's a circle of shit we're in the middle of it
but soon we'll all be dead

Like everybody else that ever lived before, the things you make fall to the floor
and nobody knows how hard you tried, it's been that way since the start of time-
There was a caveman that did some amazing things
but nobody here gives a fuck
And in a thousand years they'll feel the same
towards all the things you've done

[Chorus:]
So don't worry what might give you cancer-
or stay up nights just wanting answers
Its just a crap shoot, but it's mostly crap
things start off they're so terrific
they'll fuck up it's scientific
Entropy, uncertainty won't yield to you

Love at first sight on a beautiful night and a feeling so divine
gets sucked down the toilet cuz something will spoil it
with the good times left behind you
Don't try to figure out who's at fault
powerful forces abound
like a twig on a river in the universe
tomorrow you'll probably drown

[Chorus:]
So don't worry if it is a tumor -
all this will be over sooner
Its just a crap shoot, but it's mostly crap
things start off they're so terrific
they'll fuck up it's scientific
Entropy, uncertainty won't yield to you or you

And all the things that matter most
disappear, here's a toast
to erosion and corrosion, Altimzers and pain

Don't try to figure out who's at fault
there's powerful forces at play
and if you lose your legs and have to beg
it's really all the same

[Chorus:]
So don't worry what might give you cancer-
or stay up nights just wanting answers
Its just a crap shoot, but it's mostly crap
things start off they're so terrific
they'll fuck up it's scientific
Entropy, uncertainty won't yield to you, or you

Re:Existing

[Ignorant Aardvark]

Which part is over-estimated? All I can speak on from experience is AntiVandalBot. I ran that on an Athlon XP 2500+ (which wasn't particularly amazing at the time). It wasn't the computation that was hard, it was the network usage of downloading the diff of every edit by a non-trusted user from the RC feed. I would not have been able to run it on any home Internet connection. Thankfully I was able to place my server on an unthrottled 100 Mbps dorm connection at the University of Maryland.

I will grant you that highspeed Internet access has become a lot more widespread since 2006 (I personally have 25/15 FIOS), but at the time, there wasn't anything available residentially that could handle it.