Computer program understands the 'that’s what she said' joke

( -- While computers can do just about anything these days, having a sense of humor is not something they have been capable of, that is until now. Chloe Kiddon and Yuriy Brun, computer scientists from the University of Washington, have created a software program capable of giving computers a sense of humor and the ability to understand a specific type of double entendre.

The "that's what she said" jokes happen when a simple and innocent sentence can become a dirty joke with the addition of those four little words at the end. The example they used was the sentence "Don't you think these buns are a little too big for this meat?"

In order to create the program, the researchers began by analyzing text, including 1.5 million erotic sentences and 57,000 standard literature sentences. They looked at adjectives, nouns and verbs with high “sexiness” ratings such as “hot” and "meat."

The new system, known as DEviaNT (Double Entendre via Noun Transfer), rates sentences by looking at words that can be interpreted in more than one way and the possibility that they would fit with the “that’s what she said” scenario.

Using jokes they had gathered from online websites, they trained DEviaNT and received around 72% accuracy, though they believe they could increase that to 99.5%. They plan to present DEviaNT at the Annual Meeting of the Association for Computational Linguistics in June and plan to continue their work and expand DEviaNT's joking ability.

More information: Conference paper: THAT'S WHAT SHE SAID: DOUBLE ENTENDRE IDENTIFICATION, by Chloe Kiddon and Yuriy Brun:

© 2010

Citation: Computer program understands the 'that’s what she said' joke (2011, May 2) retrieved 9 December 2023 from
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.

Explore further

How listeners perceive verbs


Feedback to editors