-
Notifications
You must be signed in to change notification settings - Fork 16
Technical Paper
We propose a computer humor competition as a method for increasing the overall standard of computer generated jokes. We propose a challenge similar to Robocup’s 2050 FIFA World Cup champion challenge, specifically to have a computational humor system beat a human standup by 2020 in front of a live human audience. In the first instance the competition will be an annual computer versus computer contest with human judges, similar in format to the Loebner prize, or chatbot challenge. We have sought and obtained crowdsourced funding to offer a small monetary prize to the winner of the annual competition and are working to facilitate open source projects to make entrance into the competition as easy as possible.
In the last 20 years there have been a number of computational systems that either detect or generate humor. Many of them have been related to puns (Binsted & Ritchie, 1994) although recently there have been some interesting developments in double entendre detection (Kiddon & Brun, 2011). Some of these systems have been reasonably humorous; however few would argue that computers are able to generate laughter amongst humans on a regular basis.
As one means of helping increase the overall standard of computer generated jokes we propose a computer humor competition along the lines of the successful Robocup competition. We propose a challenge similar to Robocup’s 2050 FIFA World Cup champion challenge, which the Robocup web site states as follows:
By mid-21st century, a team of fully autonomous humanoid robot soccer players shall win the soccer game, comply with the official rule of the FIFA, against the winner of the most recent World Cup. Grammatical errors not-with-standing Robocup has galvanized the AI Robotics field and, since its first official games and conference in 1997, there has been clear progress in the field of robotic soccer (Kitano, 1997).
We propose a challenge of having a computational humor system be judged as funny or funnier as a professional human standup in front of a live audience by 2020. Knight (2011) has already demonstrated the feasibility of robot standup and we feel that an annual competition will serve to galvanise research and development in this area. In the first instance the competition will be an annual computer versus computer contest with human judges, similar in format to the Loebner prize (URL1), or chatbot battles (URL2). We have sought and obtained crowdsourced funding (URL3) to offer a small monetary prize to the winner of the annual competition and are releasing open source computational humor projects (URL4) to make entrance into the competition as easy as possible.
- Binsted, K. & Ritchie G. (1994) An implemented model of punning riddles. National Conference on Artificial Intelligence (AAAI), Seattle, Washington.
- Kiddon C. & Brun Y. (2011). That’s What She Said: Double Entendre Identification. In Proc. 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, 89–94.
- Kitano, H. (1998) RoboCup-97: Robot Soccer World Cup I, Springer-Verlag, LNCS 1395.
- Knight, H. (2011) Eight Lessons Learned about Non-verbal Interactions through Robot Theater. Springer-Verlag LNCS 7072, 42-51.
- URL1: http://www.loebner.net/Prizef/loebner-prize.html
- URL2: http://www.chatbotbattles.com/
- URL3: http://www.indiegogo.com/Funniest-Computer-Ever
- URL4: https://github.com/tansaku/twss