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 and non-TWSS text from sites such as
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."

Reader Comments

Subscribe: RSS

View by: Time | Thread

  1. identicon
    Josh, 30 Apr 2011 @ 12:30am

    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.

Add Your Comment

Have a Techdirt Account? Sign in now. Want one? Register here
Get Techdirt’s Daily Email
Use markdown for basic formatting. HTML is no longer supported.
  Save me a cookie
Follow Techdirt
Insider Shop - Show Your Support!

Report this ad  |  Hide Techdirt ads
Essential Reading
Techdirt Deals
Report this ad  |  Hide Techdirt ads
Techdirt Insider Chat
Report this ad  |  Hide Techdirt ads
Recent Stories
Report this ad  |  Hide Techdirt ads


Email This

This feature is only available to registered users. Register or sign in to use it.