To teach the program to spot jokes, the researchers first gave it a database of words, extracted from a children’s dictionary to keep things simple, and then supplied examples of how words can be related to one another in different ways to create different meanings. When presented with a new passage, the program uses that knowledge to work out how those new words relate to each other and what they likely mean. When it finds a word that doesn’t seem to fit with its surroundings, it searches a digital pronunciation guide for similar-sounding words. If any of those words fits in better with the rest of the sentence, it flags the passage as a joke. The result is a bot that “gets” jokes that turn on a simple pun.

Taylor presented the bot at the American Association for Artificial Intelligence conference in Vancouver, Canada, last week but stresses that it does still miss some puns. And of course, there are many jokes that aren’t based on puns, which the bot doesn’t get. Taylor notes that past experiences are often the key to why some people find things hilarious when others don’t. “If you’ve been in a car accident, you probably won’t find a joke about a car accident funny,” she says. She is now working to personalise the bot’s sense of humour by flagging certain links between words as either funny or not, depending on the experiences of people it might converse with. Meanwhile Rada Mihalcea and colleagues at the University of North Texas in Denton have built a different kind of humour-spotting bot. Instead of working out why a sentence might be funny, it learns the frequencies of words that are found in jokes, and uses that to identify humour.

“We got a lot of ‘can’t’, ‘don’t’, ‘drunk’ and ‘poor’,” Mihalcea says. “People like laughing about bad things.”

"This article is posted on this site to give advance access to other authorised media who may wish to report on this story, or quote extracts as part of fair dealing with this copyrighted material. Full attribution is required, and if reporting online a link to www.newscientist.com is also required. This story posted here is the EXACT text used in New Scientist magazine, therefore advance permission is required before any and every reproduction of each article in full. Please contact claire.bowles@rbi.co.uk. Please note that all material is copyright of Reed Business Information Limited and we reserve the right to take such action as we consider appropriate to protect such copyright."

THIS ARTICLE APPEARS IN NEW SCIENTIST MAGAZINE ISSUE: 4 AUGUST 2007. EMBARGOED UNTIL WEDNESDAY 1 AUGUST 2007, 13:00 HRS ET US.

EDITOR’S NOTE: PRIOR PERMISSION IS REQUIRED BEFORE ANY REPRODUCTION OF THIS STORY IN FULL

Author: Michael Reilly, New Scientist San Francisco office

IF REPORTING ON THIS STORY, PLEASE MENTION NEW SCIENTIST AS THE SOURCE AND, IF REPORTING ONLINE, PLEASE CARRY A LINK TO: http://www.newscientist.com

UK CONTACT - Claire Bowles, New Scientist Press Office, London: Tel: +44(0)20 7611 1210 or email claire.bowles@rbi.co.uk

US CONTACT – New Scientist Boston office: Tel: +1 617 386 2190 or email jill.heselton@reedbusiness.com