home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.ai
- Path: sparky!uunet!microsoft!wingnut!raja
- From: raja@microsoft.com (Rajasekhar Abburi)
- Subject: Re: Loebner Turing test for Human like conversation
- Message-ID: <1992Nov09.200304.20368@microsoft.com>
- Date: 09 Nov 92 20:03:04 GMT
- Organization: Microsoft Corp.
- Keywords: Loebner, turing, natural language
- Summary: Loebner prize
- References: <1992Nov5.212849.7986@cs.ucla.edu>
- Lines: 35
-
- In article <1992Nov5.212849.7986@cs.ucla.edu>, colby@oahu.cs.ucla.edu (Kenneth Colby) writes:
- > In the summer issue of AI Magazine, there is an article on the Loebner
- > Prize competition for most human-like computer conversationalist, wriiten by
- > Robert Epstein, director Cambridge Center for Behaviorial Studies.
- >
- > The competition involves Turing-like tests in using ordinary natural language
- > in a specific domain. Daniel C. Dennett is chairman of the prize committee.
- > The next competition will be held 12/15/92.
- >
- > Does anyone have any comments, criticisms, or suggestions about the article
- > or the competition.
- >
- > Ken Colby - Malibu Artifactual Intelligence Works
-
-
-
- I think that Robert Epstein's article is fairly objective and well written.
- The fundamental approach to the Loebner prize contest is resonable.
- However the technical difficulties changed the course of the 1991 contest.
- With the two failed terminals, the originally set balance was lost (with
- 6 computer programs and only 2 human "confederates"). The 2 human confeds
- may not have been as representative of the diverse spectrum of humanness
- as the originally planned 4 would have been.
-
- Most importantly, I believe that one common topic should be selected for
- all the contestants. If each contestant has a separate topic, they get
- differing latitudes for incomprehension and incorrect responses.
- Whimsical conversation, (the topic of the winning program, PC Therapist III),
- for example, can get away with random responses more easily than
- the other programs could.
-
- I think that Loebner prize contest is a good incentive for
- non-mainstream AI programmers.
-
- Raja
-