[ai_group] FW: Re: What's everyone working on?
- From: "Dr.X" <drx@xxxxxxx>
- To: "Ai_Group@Freelists. Org" <ai_group@xxxxxxxxxxxxx>
- Date: Fri, 12 Apr 2002 01:33:01 -0400
Ok, maybe we should try something simple. Just to get our feet wet. It's too
late to get into the chatterbox challenge for this round, but maybe we can on
the next. I say we learn how to deal with syntax and simple conversation
problems first and win the next chatterbox challenge. Meantime, it will help us
to figure out some of the logic problems that our ai will encounter. Things like
someone referring to something previously mentioned in the conversation, using
the word " IT " in reference to something 3 or 4 sentences back. That would
actually be impressive after reading some of the logs at the contest now. (we
would rock ;-)
Dr.X
PS: did you guys check out the lexical database link I posted? It's big, I know,
but if you would rather get it on CD, let me know. I'll see what I can do.
-----Original Message-----
From: ai_group-bounce@xxxxxxxxxxxxx [mailto:ai_group-bounce@xxxxxxxxxxxxx]On
Behalf Of David Garner
Sent: Thursday, April 11, 2002 8:22 PM
To: ai_group@xxxxxxxxxxxxx
Subject: [ai_group] Re: What's everyone working on?
John,
Sounds like you've got some experience we could learn from.
Might be one pro here.
Based on your comments, sounds like you have a feel for something simple
to try?
I'm looking for a way to get started. I've not done anything in this
area and looking to see something really simple that might shed some
light for me.
I've dug out a book I've had on the shelves from 1985 to try and get
some concepts. I've also searched the net for some simple source to
start from.
I'd like to have a simple project to code up in BCX. I think it has the
speed and flexibility to accomplish anything we might want (can always
drop to C or ASM for any speed critical stuff).
Dr.X has been trying to get me thinking in the right mind set, but I'm
still not to the point where I think I have anything to attempt
prototyping.
Any suggestions for a really simple proof of concept? I gather that
there are lots of ideas that I've not encountered that are required to
get very far in this field. I like to get a handle on the simple
concepts before trying to do so much that I don't end up with anything.
David Garner
Other related posts:
