[comp.ai] Conversations

dailey@frith.uucp (Chris Dailey) (01/31/91)

I'm not an AI person (I'm just an undergrad studying compilers), and I
have a couple of ideas regarding conversation in natural languages.

I wanted to ask if anyone here knew of any work done in this area.  The
only thing even close to what I'm thinking of is the ancient ELIZA-type
programs.  Some of the more complex adventure games may also come close
to what I'm thinking of, but probably not.

I guess what I'm wondering about is what is being done on getting a
computer to follow a logical conversation, whether that be in the user
querying the computer or the computer querying the user.

I just checked our library computer here at MSU (and UofM's, too) WRT
Conversations, and this is what it said:

Studies in the organization of conversational interaction / edited by Jim
  Schenkein. -- New York : Academic Press, c1978.
  xvi, 275 p. ; 24 cm. -- (Language, thought, and culture)
  Bibliography: p. 271-275.
 SUBJECT HEADINGS (Library of Congress; use s= ):
     Interpersonal communication.
     Conversation--Research.

I have not yet been able to check this out, but I would suspect it would
have little to do with computer application.  There was nothing under
ARTIFICIAL INTELLIGENCE that seemed what I was looking for, either.

I guess what I'm interested in is stream of consciousness.  If you can
give me a simplified description of a theory or concept, that would be
great -- pointers to sources of information would be acceptable, also.
I have redirected follow-ups to myself, and if people request, I will
summarize.

Thanks,
--
Chris Dailey   dailey@(frith.egr|cps).msu.edu
    __  __  ___       | "A line in the sand." -- The Detroit News
 __/  \/  \/ __:>-    |
 \__/\__/\__/         | "Allein in der sand." -- me

andrew@calvin.doc.ca (Andrew Patrick) (02/02/91)

In article <1991Jan30.164108.24633@msuinfo.cl.msu.edu> dailey@frith.uucp (Chris Dailey) writes:
>I'm not an AI person (I'm just an undergrad studying compilers), and I
>have a couple of ideas regarding conversation in natural languages.
>
>I wanted to ask if anyone here knew of any work done in this area.  The
>only thing even close to what I'm thinking of is the ancient ELIZA-type
>programs.  Some of the more complex adventure games may also come close
>to what I'm thinking of, but probably not.
...

We have published a couple of papers on this topic recently:

Whalen, T. E., & Patrick, A. S.  (1990).  COMODA:  A conversation model
    for database access.  Behaviour & Information Technology, 9, 93-110.

Whalen, T.E, & Patrick, A.S.  (1989).  Conversational hypertext:
    Information access through natural language dialogues with
    computers.  Paper presented at the SIGCHI Conference, April 30 - May
    4, 1989, Austin Texas.  Published in CHI '89 Conference Proceedings,
    289-292.

We also have some unpublished papers you may be interested in -- drop
me a note with a postal address.


    BTW, my news software (the latest C-News and NNTP) coughs with
    follow-ups directed to an e-mail address -- it complains that there
    are not newsgroups matching foo@bar.domain in the active file (which
    makes sense).  Is this practice of directing followups to an e-mail
    address common?


-- 
Andrew Patrick, Ph.D.       Department of Communications, Ottawa, CANADA
andrew@calvin.doc.CA
                    "The interface IS the program."