Re: Problem using simple fuzzy logic in artifical intelligence


Subject: Re: Problem using simple fuzzy logic in artifical intelligence
From: Valeri Sitnikov (Valerij.Sitnikov@mow.siemens.ru)
Date: Fri Oct 20 2000 - 11:17:27 MET DST


I would say, use analogy ;)
Actually what I have found that it's much better sometimes to reformulate
the task not in terms of Fuzzy sets but in terms of similarity (I have a
paper on it, if you are interested). So, what you do is that you reformulate
the task and use a function (it can be any in principle) of similarity, f.e.

In order it to be TimeLeft(lot)=0.5 and this is true, you just need to have
a function which is

f(x,0.5) = 1

and in any other place it shows how similar is the arguments to 0.5.

Here are some functions one can use:

f(x,y)= exp(-(x-y)*(x-y))
f(x,y)= min(x,y)/max(x,y)

I have found out that whatever the function is it doesn't harm the logic if
it represents Analogy.
If you like to have more detailed descriptions you can get the .doc file
from
http://members.fortunecity.com/vvssvv/

If any questions, get in touch.

Valeri

> Now, the problem:
> If the crisp (sharp?) time is e.g. 10 minutes left then what is the fuzzy
> time_left(lot)? If I say that time_left(lot) is true when
> TimeLeft(lot)>=0.5 then is it fuzzy anymore??? It means that at some point
> of the match there is a lot of time left but the next minute only some
> time left. I could construct it without any fuzzy rules using plain
> integer values. Is there any simple method that I could use or is this a
> wrong place to use fuzziness anyway?
>
> I appologize the fuzziness of explanation, I hope that someone got the
> point.
>
>
> Teemu Hiltunen
> teemu.hiltunen@uta.fi

############################################################################
This message was posted through the fuzzy mailing list.
(1) To subscribe to this mailing list, send a message body of
"SUB FUZZY-MAIL myFirstName mySurname" to listproc@dbai.tuwien.ac.at
(2) To unsubscribe from this mailing list, send a message body of
"UNSUB FUZZY-MAIL" or "UNSUB FUZZY-MAIL yoursubscription@email.address.com"
to listproc@dbai.tuwien.ac.at
(3) To reach the human who maintains the list, send mail to
fuzzy-owner@dbai.tuwien.ac.at
(4) WWW access and other information on Fuzzy Sets and Logic see
http://www.dbai.tuwien.ac.at/ftp/mlowner/fuzzy-mail.info
(5) WWW archive: http://www.dbai.tuwien.ac.at/marchives/fuzzy-mail/index.html



This archive was generated by hypermail 2b25 : Fri Oct 20 2000 - 11:19:20 MET DST