Newsgroups: comp.ai
Path: cantaloupe.srv.cs.cmu.edu!das-news.harvard.edu!news2.near.net!MathWorks.Com!europa.eng.gtefsd.com!howland.reston.ans.net!cs.utexas.edu!uunet!brunix!doorknob.cs.brown.edu!jgm
From: jgm@tahoe.cs.brownie.edu (Jonathan Monsarrat)
Subject: What came after ELIZA?
Message-ID: <JGM.94Sep2141127@tahoe.cs.brownie.edu>
Sender: news@cs.brown.edu
Organization: Brown University Department of Computer Science
Date: Fri, 2 Sep 1994 18:11:27 GMT
Lines: 30

Hi!

There are a lot of on-line games where you encounter "The Sheriff",
an AI construct in the game, and you and "The Sheriff" have a limited
conversation. This can be done in many ways:

    1. The Sheriff searches for keywords in your sentences, like ELIZA.
    2. You are given a choice of 5 things to say. You choose one, The
       Sheriff responds, and then you get a choice of 5 more things.
       An expert system controls The Sheriff's response for all
       combinations of things you say.
    3. You take actions like

       -- give sheriff the wand
       -- punch sheriff
       -- bow to sheriff

       and because taking actions has a limited vocabulary, it's
       easier for The Sheriff to react.

I'm wondering what more can be done with such autonomous agents.

I'm sure there's a word for this kind of AI research, but I don't know it.

Any references? :)

-Jon
%! Jon Monsarrat     jgm@cs.brown.edu     moderator, comp.sources.postscript %!
9 9 scale 9 9 moveto(qll-??LHHL??llH?hH7t,7olCAHH@){dup 10 mul rotate 80 lt{4 0
rlineto}{4 0 rmoveto}ifelse}forall stroke showpage
