Talk:Fuzzy set operations

From Wikipedia, the free encyclopedia

Mathematics Portal

This article is within the scope of WikiProject Mathematics, which collaborates on articles related to mathematics.

Mathematics rating:

Stub Class

Low Priority

Field: Foundations, logic, and set theory

Please update this rating as the article progresses, or if the rating is inaccurate. Please also add comments to suggest improvements to the article.

should the subidempotency axiom for the intersection be "i(a,a)<= a" rather than "i(a,a) < a" ?

Although in mathematics the prefix 'sub-' is usually understood in a non-strict way (a set is a subset of itself, a group is a subgroup of itself, and so on) in this case the property "i(a,a)<=a" already follows from the other axioms, since "i(a,a)<=i(a,1)=a", so only strict subidempotency is a non-trivial property. It is standard to call the property "i(a,a)<a" either subidempotency (as done in the Klir-Yuan textbook) or Archimedeanity, however it must be noted that it is not equivalent to the ordinary meaning of the Archimedean property (i.e. the iteration of any non-trivial "a" by the operation eventually yields a number as close to 0 as wished) unless further assumptions on "i" are imposed (e.g. continuity). Of course, for "u" the above should read `as close to 1 as wished'.

I think it would be important to note which of the axioms are `true' axioms (those defining a triangular norm and a triangular conorm) and which are in fact additional requierements which may or may not be imposed. Otherwise this mixed usage of the word `axiom' will be confusing to any unaware reader. For instance, the minimum is called the standard intersection but it fails the subidempotency `axiom'.--155.210.235.125 17:50, 5 July 2006 (UTC)

[edit] Other fuzzy set operations

Here's what I found in the book I have (Pospelov, 1986)

Absolute complement (NOT): $1 - \mu_1 \left( x \right)$
Intersection (AND) 1 (Standard intersection): $\min \left( \mu_1 \left( x \right), \mu_2 \left( x \right) \right)$
Union (OR) 1 (Standard union): $\max \left( \mu_1 \left( x \right), \mu_2 \left( x \right) \right)$
Intersection (AND) 2 (Bounded difference): $\max \left( 0, \mu_1 \left( x \right) + \mu_2 \left( x \right) - 1 \right)$
Union (OR) 2 (Bounded sum): $\min \left( 1, \mu_1 \left( x \right) + \mu_2 \left( x \right) \right)$
Intersection (AND) 3 (Algebraic product): $\mu_1 \left( x \right) \times \mu_2 \left( x \right)$
Union (OR) 3 (Algebraic sum): $\mu_1 \left( x \right) + \mu_2 \left( x \right) - \mu_1 \left( x \right) \times \mu_2 \left( x \right)$
Relative Complement: $\max \left( 0, \mu_1 \left( x \right) - \mu_2 \left( x \right) \right)$
Concenration (VERY): $\mu_1 \left( x \right) ^ 2$

In other references I saw one more:

Deconcenration (MORE OR LESS): $\sqrt {\mu_1 \left( x \right) }$

I'm not sure if this is accurate as it seems English and Russian sources differ on this topic. Solarapex 11:35, 30 July 2007 (UTC)

Added corresponding terms I found in English references.

A couple more:

Drastic intersection: $\begin{cases} \mu_1, & \mu_2 = 1 \\ \mu_2, & \mu_1 = 1 \\ 0, otherwise \end{cases}$
Drastic union: $\begin{cases} \mu_1, & \mu_2 = 0 \\ \mu_2, & \mu_1 = 0 \\ 1, otherwise \end{cases}$