Scientists Finally Tackle The Age Old 'That's What She Said' Problem
from the it-was-hard-and-it-took-forever dept
Forget the traveling salesman problem or p=np, some computer scientists have finally tackled the really big challenge for computers: teaching them how to understand the innuendo and double entendres necessary to make a “that’s what she said” joke. Yes, all other computer science pale in comparison, so kudos to Chloe Kiddon and Yuriy Brun for tackling such a difficult challenge:
Automating this process means identifying sentences that contain potential euphemisms and follow a particular structure – a “hard natural language understanding problem”, say the researchers. Kiddon and Brun began by analysing two different bodies of text – one containing 1.5 million erotic sentences, and another with 57,000 from standard literature.
They then evaluated nouns, adjectives and verbs with a “sexiness” function to determine whether a sentence is a potential TWSS. Examples of nouns with a high sexiness function are “rod” and “meat”, while raunchy adjectives are “hot” and “wet”.
Their automated system, known as Double Entendre via Noun Transfer or DEviaNT, rates sentences for their TWSS potential by looking for particular elements such as nouns that can be interpreted in multiple ways. The researchers trained DEviaNT by gathering jokes from twssstories.com and non-TWSS text from sites such as wikiquote.org.
Apparently, the system is about 70% accurate so far, but they believe they can get it up to 99.5% accuracy before too long.
I’m sorry, Watson, but this may be the biggest computing/artificial intelligence story of the year. And, already, the race is on to come up with the appropriate jokes. My favorite so far was this quote for the researchers on this project: “It was hard and it took forever.”
Filed Under: computers, double entendre, jokes, that's what she said
Comments on “Scientists Finally Tackle The Age Old 'That's What She Said' Problem”
Another trick would be to identify ones where someone might say “That’s what she said.” but there’s absolutely no innuendo. Anyone might just stare back with a wrinkled forehead saying “What…? That makes no sense”. I wonder how the program would react.
Re: Failed ones?
That’s what he said.
Re: Failed ones?
“I’m sorry, Watson, but this may be the biggest computing/artificial intelligence story of the year.”
That’s what she said.
I can’t do that Dave. THAT’s what she said.
Even though this story seems kind of silly on the surface, it wouldn’t surprize me if this was the kind of work that leads to the true revolutionary breakthroughs in AI research.
I can’t even imagine where to begin teaching a machine to recognize euphemisms, innuendo, allusions and metaphors.
Soon, we will be able to automate the production of great art and literature, freeing up a lot of time for us to hang out with the sentient robots and make dirty jokes.
You guys think innuendo is all about “noun transfer”?
Better luck next time.
“Better luck next time”… I bet you hear her say that all the time!
I wonder what a “sexy function” would look like in a line of code.
var sexy = lulz;
if (findfunny == sexy)
alert(“That’s what she said!”);
else if (findfunny /= sexy)
There’s a bug in your code. ‘Findfunny’ is undefined in the evaluation of the conditional… and that’s what she said.
Re: Re: Re:
Its not undefined, Its just going to pull some random value left over in memory from some other app essentially making it if( math.random() == sexy ) which will probably be just about as accurate as the actual program.
While impressive, this sort of AI is imitative rather than innovative. Imagine if the resources it pulled from (which were created by humans) weren’t there or if there were no immediate pattern it could “learn” from. This isn’t an understanding of innuendo, it’s an understanding of a particular pattern. Humans don’t need to be taught how to appreciate a TWSS joke by listening to a bunch of other TWSS jokes first. If they did, the jokes would have never been spawned (chicken vs. egg conundrum). This is an example of “studying for the Turing Test” rather than creating an AI that by its nature is able to pass the Turing Test, not that the former isn’t still an achievement.
Not so fast. (This is too easy!)
This release was obviously premature.
I never thought you would announce this to the whole world.
I saw this yesterday but didn’t think much of it.
You are way off the mark.
You are so far off it hurts.
Do you think I’m done yet?
You’re working so hard at this — it’s so funny, it’s almost touching!
What else do you think you can do with this little achievement?
You’re looking in the wrong place.
What you’re calling a “function” is way too mechanical.
Don’t force it.
I’m not really done, but let’s just leave it there for a while.
Is this really something that needs investigating? What do you think will come of it?
That article was long and hard.
I for one welcome our “that’s what she said” responding terminator overlords!
Double entendres and computers
After some really excellent analyses (such as the ridiculous and progress-inhibiting class action law suits), you really blew it on this one.
There are very few things more important than providing a means for understanding subtleties in computers. Otherwise, when our (potential) help-mates encounter “I wish I were dead” from a teenager with a history test, the “help” provided will be a call to 911!
This is important stuff. I am amazed you don’t see it.
At the very least Congress can embed the system into cellphones and finally have a sure fire way to ruin sexting for teenagers.