MEMBERSHIP

FUNCTIONS
 MEMBERSHIP FUNCTIONS

• Define the fuzziness in a fuzzy set irrespective of the elements in the set, which are
discrete or continuous.
• The membership functions are generally represented in graphical form.
• The rules that describe fuzziness graphically are also fuzzy.
• Membership function can be thought of as a technique to solve empirical problems on
the basis of experience rather than knowledge.
• Available histograms and other probability information can also help in constructing
the membership function.
• There are several ways to characterize fuzziness; In a similar way, there are several
• ways to graphically construct a membership function that describes fuzziness.
• We will deal with a few possibilities of describing membership functions.
CRISP MEMBERSHIP FUNCTIONS

Crisp membership functions () are either one or zero.

Consider the example: Numbers greater than 10. The membership

curve for the set A is given by
REPRESENTING A DOMAIN IN FUZZY LOGIC
FUZZY MEMBERSHIP FUNCTIONS
The set B of numbers approaching 2 can be represented by the
membership function
FUZZINESS vs PROBABILITY
LINGUISTIC VARIABLE
LINGUISTIC VARIABLE

Let xbe a linguistic variable withthe label“speed”.

Terms of x, which are fuzzy sets, could be “positive low”, “negative
high” from the term setT:

T = {PostiveHigh, PositiveLow, NegativeLow,

NegativeHigh, Zero}

Each term is a fuzzy variable defined on the base variable which

might be the scale of all relevant velocities.
MEMBERSHIP FUNCTIONS
FEATURES OF MEMBERSHIP FUNCTIONS

• CORE:

• SUPPORT:

• BOUNDARY:
 FEATURES OF MEMBERSHIP FUNCTIONS
Core: The core of a membership function for some fuzzy set A is defined as that region of
universe that is characterized by complete membership in the set A. The core has
elements x of the universe such that:

The core of a fuzzy set may be an empty set.

Support: The support of a membership function for a fuzzy set A is defined as that region
of universe that is characterized by a non zero membership in the set A. The support
comprises elements x of the universe such that:

A fuzzy set whose support is a single element in X with is referred to as a fuzzy

singleton.

Boundary: The boundary of a membership function for a fuzzy set A is defined as that
region of universe containing elements that have a non zero but not complete
membership. The boundary comprises those elements of x of the universe such that:

The boundary elements are those which possess partial membership in the fuzzy set A
 FEATURES OF MEMBERSHIP FUNCTIONS

• A fuzzy set whose membership function has at least one element x in the universe
whose membership value is unity is called normal fuzzy set.
• The element for which the membership is equal to 1 is called prototypical element.
• A fuzzy set wherein no membership function has its value equal to 1 is called
subnormal fuzzy set.

Normal Fuzzy Set Subnormal Fuzzy Set

 FEATURES OF MEMBERSHIP FUNCTIONS

• A convex fuzzy set has a membership function whose membership values are strictly
monotonically increasing or strictly monotonically decreasing or strictly monotonically
increasing than strictly monotonically decreasing with increasing values for elements in
the universe.
• A fuzzy set possessing characteristics opposite to that of convex fuzzy set is called
nonconvex fuzzy set, i.e., the membership values of the membership function are not
strictly monotonically increasing or decreasing or strictly monotonically increasing than
decreasing.

Convex Normal Fuzzy Set Nonconvex Normal Fuzzy Set

 FEATURES OF MEMBERSHIP FUNCTIONS

• For elements x1, x2 and x3 in a fuzzy set A, if the following relation between x1, x2 and x3,
holds. i.e.
then A is said to be a convex fuzzy set.
• The membership of the element x2 should be greater than or equal to the membership
of elements x1 and x3· For a nonconvex fuzzy set, this constraint is not satisfied.
• The intersection of two convex fuzzy sets is also a convex fuzzy set.
• The element in the universe for which a particular fuzzy set A has its value equal to 0.5
is called crossover point of a membership function.
• The membership value of a crossover point of a fuzzy set is equal to 0.5. i.e., μA(x) = 0.5.
• There can be more than one crossover point in a fuzzy set.
• The maximum value of the membership function in a fuzzy set A is called as the height
of the fuzzy set.
• For a normal fuzzy set, the height is equal to 1.
• Thus, if the height of a fuzzy set is less than 1, then the fuzzy set is called subnormal
fuzzy set.
• When the fuzzy set A is a convex single-point normal fuzzy set defined on the real time,
then A is termed as a fuzzy number.
FUZZIFICATION
Fuzzification is the process of transforming a crisp set to a fuzzy set or a fuzzy
set to a fuzzier set, i.e., crisp quantities are converted to fuzzy quantities.
This operation translates accurate crisp input values to linguistic variables.
 FUZZIFICATION

• For a fuzzy set A = {μi/xi | xi ϵ X}, a common fuzzification algorithm

is performed by keeping μi constant and xi being transformed to
a fuzzy set Q(xi) depicting the expression about xi. The fuzzy set
Q(xi) is referred to as the kernel of fuzzification. The fuzzified set
A can be expressed as:
• A = μ1 Q(x1) + μ2 Q(x2) + ….. + μn Q(xn)
• This process of fuzzification is called support fuzzification (s-
fuzzification).
• There is another method of fuzzification called grade fuzzification
(g-fuzzification) where xi is kept constant and μi is expressed as a
fuzzy set.
• Thus, using these methods, fuzzification is carried out.
• Use crisp inputs from the user.

• Determine membership values for all the relevant classes

(i.e., in right Universe of Discourse).
EXAMPLE - FUZZIFICATION
FUZZIFICATION OF HEIGHT
FUZZIFICATION OF WEIGHT
 METHODS OF MEMBERSHIP VALUE
ASSIGNMENT
The various methods of assigning membership values are:

• Intuition,
• Inference,
• Rank ordering,
• Angular fuzzy sets,
• Neural networks,
• Genetic algorithm,
• Inductive reasoning.
 METHODS OF MEMBERSHIP VALUE
ASSIGNMENT
• Intuition: lt is the capacity of the human to develop membership functions on the basis
of their own intelligence and understanding capability. There should be an in-depth
knowledge of the application to which membership value assignment has to be made.
• Inference: The inference method uses knowledge to perform deductive reasoning.
Deduction achieves conclusion by means of forward inference.
• Here the knowledge of geometrical shapes and geometry is used for defining
membership values.
• The membership functions may be defined by various shapes: triangular,
trapezoidal, bell-shaped, Gaussian and so on.
• Consider a triangle, where X, Y and Z are the angles, such that X ≥ Y ≥ Z ≥ 0, and let
U be the universe of triangles, i.e.,
U = [(X,Y,Z) | X ≥ Y ≥ Z ≥ 0; X + Y + Z = 180]

Let us consider the following triangles:

I = isosceles triangle (approximate)
E = equilateral triangle (approximate)
R = right-angle triangle (approximate)
IR = isosceles and right-angle triangle (approximate)
T = other triangles
 METHODS OF MEMBERSHIP VALUE
ASSIGNMENT
The membership values of approximate isosceles triangle is obtained using the following
definition, where:
X ≥ Y ≥ Z ≥ 0 and X+ Y+ Z = 180°:
μI(X,Y,Z) = 1 – 1/60° min(X – Y, Y - Z)
If X = Y or Y = Z, the membership value of approximate isosceles triangle is equal to 1. On
the other hand if X = 120°, Y = 60° and Z = 0° we get:

The membership value of approximate right-angle triangle is given by:

μg(X,Y,Z) = 1 – 1/90°(X – 90°)
For X = 90° ➔ μg = 1
For X = 180° ➔ μg = 0
 METHODS OF MEMBERSHIP VALUE
ASSIGNMENT
The membership value of approximate isosceles righr-angle triangle is obtained by taking
the logical intersection of the approximate isosceles and approximate right-angle triangle
membership function i.e., IR = I ∩ R
μIR(X,Y,Z) = min[μI(X,Y,Z), μR(X,Y,Z)]
= 1 – max[1/60° min(X – Y, Y - Z), 1/90°(X – 90°)]
The membership function for a fuzzy equilateral triangle is given by
μE(X,Y,Z) = 1 – 1/180°(X – Z)
The membership function of other triangles, denoted by T, is the complement of the
logical union of I, R and E i.e.
T = I∪R∪E
By using De Morgan’s Law we get:
T = 𝐼∩𝑅∩𝐸
The membership value can be obtained by using the equation:
μT(X,Y,Z) = min[1 - μI(X,Y,Z), 1 - μE(X,Y,Z), 1 - μR(X,Y,Z)]
= 1/180° min[3(X – Y), 3(Y - Z), 2(X – 90°), X - Z]
The inference method as discussed for triangular shape can be extended for trapezoidal
shape and so on,
on the basis of knowledge of geometry.
 METHODS OF MEMBERSHIP VALUE
ASSIGNMENT
• Rank ordering: Methodology of preferences can be adapted to assign membership
values to fuzzy variables.
• Pairwise comparisons enable us to determine preferences and this results in
determining the order of the membership.
• Angular fuzzy sets: Coordinate description is the major difference between the angular
fuzzy sets and standard fuzzy sets.
• Angular fuzzy sets are defined on a universe of angles thus repeating the shapes
every 2π cycles.
• The truth values of the linguistic variable are represented by angular fuzzy sets.
• Consider the pH value of wastewater from a dyeing industry. These pH readings
are assigned linguistic labels, such as high base, medium acid, etc., to understand
the quality of the polluted water.
• As is known, the neutral solution has a pH value of 7.
• The linguistic variables are built such that:
• “Neutral (N)” means ϴ = 0 rad
• “Exact Base (EB)” means ϴ = π/2 rad
• “Exact Acid (EA)” means ϴ = - π/2 rad
• The levels of pH between 7 and 14 can be termed as "very base" (VB),
"medium base" (MB) and so on and are represented between 0 to π/2.
 METHODS OF MEMBERSHIP VALUE
ASSIGNMENT

• Levels of pH between 0 and 7 can be termed as "very acid (VA)," "medium acid (MA)”
and so on and are represented between 0 rad and -π/2 rad

The membership value

corresponding to linguistic term
can be obtained from :
μT(ϴ) = t. tan(ϴ)
Where t is horizontal projection
of radial vector

Model of Angular Fuzzy Set

 METHODS OF MEMBERSHIP VALUE
ASSIGNMENT

• Neural Networks: The neural network can be used to obtain fuzzy

membership values. Consider a case where fuzzy membership functions are to
be created fuzzy classes of an input data set.
• The data points are first divided into different classes by conventional
clustering techniques.
• The data points are given membership values of 1 for the classes they
initially belong.
• A neural network is created which uses the data point marked 1 and the
cor-responding membership values in different classes for training itself
for simulating the relationship between coordinate locations and the
membership values.
• The process is continued until the neural network simulates the entire set
of input-output values. The network performance is tested using the
testing data set.
 METHODS OF MEMBERSHIP VALUE
ASSIGNMENT

• Genetic algorithm: Genetic algorithm is based on the Darwin's theory of evolution; the
basic rule is "survival of the finest." The genetic algorithm is used to determine the
fuzzy membership functions. This can be done using the following steps:
• For a particular functional mapping system, the same membership functions and
shapes are assumed for various fuzzy variables to be defined.
• These chosen membership functions are then coded into bit strings.
• Then these bit strings are concatenated together.
• The fitness function to be used here is noted.
• The fitness function is used to evaluate the fitness of each set of membership
functions.
• The membership functions define the functional mapping of the system.
The process of generating and evaluating strings is carried out until we get a
convergence to the solution within a generation i.e. we obtain the membership
functions with best fitness value. Thus, fuzzy membership functions can be obtained
from genetic algorithm
 METHODS OF MEMBERSHIP VALUE
ASSIGNMENT
• Induction reasoning: Induction is used to deduce causes by means of backward
inference.
• Induction employs entropy minimization principle, which clusters the parameters
corresponding to the output classes. To perform inductive reasoning method, a
well-defined database for the input-output relationship should exist.
• The inductive reasoning can be applied for complex systems where the data are
abundant and static.
• For dynamic data sets, this method is not best suited, because the membership
functions continually changes with time. There exist three laws of induction:
• Given a set of irreducible outcomes of an experiment, the induced
probabilities are those probabilities consistent with all available information
that maximize the entropy of the set.
• The induced probability of a set of independent observations is proportional
to the probability density of the induced probability of a single observation.
• The induced rule is that rule consistent with all available information of that
minimizes the entropy.
 METHODS OF MEMBERSHIP VALUE
ASSIGNMENT
The membership functions using inductive reasoning are generated as follows:
• A fuzzy threshold is to be established between classes of data.
• Using entropy minimization screening method, first determine the threshold
line.
• Then start the segmentation process.
• The segmentation process results into two classes.
• Again partitioning the first two classes one more time, we obtain three
different classes.
• The partitioning is repeated with threshold value calculations, which lead us to
partition the data set into a number of classes or fuzzy sets.
• Then on the basis of the shape, membership function is determined.
SUMMARY
The contents discussed include:

• Membership functions – Definition and properties.

• Linguisticvariables.

• Fuzzification process.