# Re: truth values if one variable is undefined?

From: Andrew Edmonds (andy@metarule.com)
Date: Sun Mar 31 2002 - 23:46:00 MEST

• Next message: Simon Garrett: "CFP ICARIS-2002"

Hello Ernst,

In our software, XmlMiner and XMLRule, which you can look at on
http://www.metarule.com we use a sort of tristate logic, although that
doesn't really sum it up. Degrees of truth are represented as values between
0 and 1 inclusive, and a "don't know" condition is coded as -1. All our
operators handle the -1 case through a truth table, and propogate the don't
know through the rule to the result. For example an "and" function
returns -1 if either input is unknown, an "or" function returns the other
input value if one of the inputs is unknown etc. For a "not" function if the
input is -1, so is the output. Any aggregation has to take care of don't
knows propogating through and respond accordingly. You can get more
information on our web site. I'd be interested to know who else uses this
system, we've used it for some time.

Best wishes,

Andy Edmonds, Scientio, Inc.

"Ernst Murnleitner" <mur@awite.de> wrote in message
news:a7mv5s\$m63j7\$1@ID-130107.news.dfncis.de...
>
> I want to handle undefined variables, i.e. if a measurement ist not
> available and thus the input value is invalid:
>
> In our software, we set a flag "invalid" to true, if the measurement is
not
> available.
> But there comes up the following question: How should the term with the
> undefined variable be handled?
> One possibility would be, to let the truth always be "0" of that sub-term,
> e.g.:
> IF A = a THEN ...
> ####
> Here, (A=a) = 0
>
> But how should NOT be handled then?
>
> IF (NOT(A=a)) THEN
> #########
> Here, (NOT(A=a)) would be "1"!
>
> Or should the complete subterm with the undefined variable be "0"? Hence,
> NOT(A=a) would be "0".
>
> (To Invalidate the whole rule, Is not a good idea: because: IF A=a OR B=b
> THEN ...)
>
> Greetings
> Ernst Murnleitner
>
>
>

############################################################################
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 2b30 : Mon Apr 01 2002 - 00:11:48 MEST