home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: sci.math
- Path: sparky!uunet!gumby!destroyer!wsu-cs!trace.eng.wayne.edu!uds
- From: uds@trace.eng.wayne.edu (Seetamraju Udaybhaskar)
- Subject: Re: Functions for n-valued logic
- Message-ID: <1992Nov12.213814.15752@cs.wayne.edu>
- Keywords: n-valued logic
- Sender: usenet@cs.wayne.edu (Usenet News)
- Reply-To: uds@trace.eng.wayne.edu (Seetamraju Udaybhaskar)
- Organization: Wayne State University, Detroit
- References: <1992Nov12.161637.20049@ghost.dsi.unimi.it>
- Distribution: world
- Date: Thu, 12 Nov 1992 21:38:14 GMT
- Lines: 17
-
- In article <1992Nov12.161637.20049@ghost.dsi.unimi.it> lombardo@ghost.dsi.unimi.it (franco lombardo) writes:
- >
- >i have a function f:{0,1/2,1}^3--->{0,1/2,1}^3 completely described
- >by a table and i want to express it using the operators of
- >a three-valued Lukasiewicz logic: and, or, not and
- >Lukasiewicz implication.
-
- The function maps its domain onto itself...
-
- Draw up a lattice, with 3^3 points. Draw the arrows, and check if
- every point can be represented as variations of the supremum
- and infremum operators... The variations will be clearly based on
- and/OR/NOT operators...
-
- This is a generalized idea for any sort of table U have...
-
- (For lattices, look up partially ordered set theory (PO-set theory))...
-