Вы находитесь на странице: 1из 6

1.

Answer:
a.

The net input for the activation can be calculated as follows:


y(net) = x1.w1 + x2.w2 + x3.w3 + . . . + xm+wm

After applying the activation function we get the output as follows:


Y = F(y(net))

b. The necessity of momentum factor in weight updation process are as follows:


i. A high learning rate (α) leads to rapid learning but the weights may
oscillate.
ii. A lower value of 'α' leads to slower learning process in weight
updating.
iii. Momentum factor (μ) is to accelerate the convergence of error during
the training.

c. The outer products rule used for training pattern association networks are as
follows:
i. The architecture can be feed-forward or recurrent.
ii. Information flows from input units to output nets in case of
feed-forward net.
iii. There are closed loops of connected nodes in case of Recurrent net.

d.
Classical Set Fuzzy Set Crisp Set

Basic Collection of Collection of infinite Collection of


distinct object object Bi-Object
Property Each individual Elements are Elements are either
entity is a allowed to be in set or not
member partially included

e. A fuzzy relation µ on X is a fuzzy equivalence relation if it is a fuzzy reflexive,


symmetric and transitive relation on X.

Example:

Let X = {a, b, c}, define λ : X × X → [0, 1] by λ (a, a) = λ (b, b) = λ (c, c) = 0.5,


λ (a, b) = λ (a, c) = 0.3, λ (b, a) = λ (b, c) = 0.4, λ (c, a) = λ (c, b) = 0.2. Then λ
is an (∈, ∈)-fuzzy reflexive relation, but not a fuzzy reflexive relation.

f. Cardinality of a fuzzy set is defined as :


i. It is the sum of the membership values of all elements of the
fuzzy set.
ii. It associates each fuzzy set to a positive real number.

No, Power sets can’t be formed for a fuzzy set.

g. The necessity of the defuzzification was arisen due to crisp control actions in
some practical implementations needed to operate the control.

h.
Genetic Algorithm Traditional Algorithm

It is based on the principle of genetic It is a step by step procedure to


and natural selection to solve solve a problem.
optimization problems.

Highly Advanced Bleh

Mainly used in Machine Learning Mainly used in boring stuff like


and cool stuff. Programming, Mathematics etc.

i. Some of the applications are as follows:


i. Forecasting
ii. Sungun Copper deposit in Northwestern Iran
iii. Real estate price.

2. Answer:
a. “AND” function in perception network is implemented as below:
def AND_func (x):
w = np.array([1,1])
b = -1.5
return perceptron(x,w,b) #neural network function

b. The architecture of BPN is:

The significance of error in BPN Algo are as follows:


1. It is a standard method for training neural networks.
2. It helps to calculate the gradient of a loss function.

5. Answer:
a. Defuzzification Method can be explained as follows:

i. It is a process of converting fuzzy quantity to crisp value.

ii. It is decided by a decision-function algorithm.

Some of the methods include are as follows:

● Centroid:

○ It returns the center of gravity of the fuzzy set along the x-axis.

● Bisector:

○ It finds the vertical line that divides the fuzzy set into two sub-regions

of equal area.

● Middle of Maximum:
○ It finds the middle of maximum plateau on the graph.

● Smallest of Maximum:

○ It finds the smallest of maximum plateau on the graph.

● Largest of Maximum:

○ It finds the largest of maximum plateau on the graph.

b. Some of the operations performed on fuzzy relations are as follows:


i. Union: The union of two relations are formed by taking the maximum (t-norm)
of the two grades of membership for the corresponding elements of the two
matrices.
ii. Intersection: The intersection of relations are formed by taking the minimum
(s-norm) of the two grades of membership for the corresponding elements of
the two matrices.
iii. Complement: The complement of a relation is formed by subtracting the
relation from 1 for the corresponding elements of the matrice.

6. Answer:
a. Properties of addition and multiplication of intervals are as follows:
i. Commutativity: A (operator) B = B (operator) A
ii. Associativity: (A (operator) B) (operator) C = A (operator) (B (operator) C)
iii. Neutral Number: A (operator) 0 = 0 (operator) A = A
iv. Image and Inverse: A + A(bar) != 0 ///////// A . A(inverse) != 1

b. A steam engine’s speed and pressure can be explained as:


“ If Speed is Slow Then Pressure should be High.”

Let x be the Speed and y be the Pressure.

In general, three forms of rule exist for any linguistic variables.


• Assignment statement
e.g. x is not large AND not very small
• Conditional statement
e.g. IF x is very big THEN y is medium
• Unconditional statement
e.g. set pressure high

Compound Rules:
Conjunctive antecedents:
For a multiple conjunctive antecedents of the form

IF x is A1 AND x is A2 … AND x is An THEN y is Bs …..(6)


can be rewritten as
IF x is As THEN y is Bs..............(7)

Where As = A1 ∩ A2 ∩ A3 ∩ … ∩ An and As is expressed by


means of membership function based on the definition of fuzzy
intersection operation as

μ(As) (x) = min[ μ(A1) (x), μ(A2) (x), …. , μ(An) (x)].........(8)

Disjunctive antecedents:
For multiple disjunctive antecedents of the form

IF x is A1 OR x is A2 … OR x is An THEN y is Bs.......(9)

can be rewritten as

IF x is As THEN y is Bs..............(10)

Where As = A1 ∪ A2 ∪ A3 ∪ … ∪ An and As

is expressed by means of membership function based on the


definition of fuzzy union operation as

μ(As) (x) = max[ μ(A1) (x), μ(A2) (x), …. , μ(An) (x)]..........(11)

Aggregation of Rules:
In the case of a system of rules that must be jointly satisfied,
the rules are connected by AND connectives. The aggregated output y
is found by fuzzy intersection of the entire individual rule consequent yi
, where i =1,2,3,L,r.

y = y 1 AND y2 AND … AND yr........(12)

Or

y = y 1 ∩ y 2 ∩ … ∩ yr

which is defined by the membership function

μ(y)(y) = min[ 1 ( y), 2 ( y), , r ( y)] for y ∈Y...................(13)

For the case of a disjunctive system of rules where at least one rule
must be satisfied, the rules are connected by OR connectives. The
aggregated output y is found by a fuzzy union of all the individual rule
consequent yi, where i =1,2,3,L,r .

y = y1 OR y2 OR … OR yr....................(14)

or

y = y 1 ∪ y 2 ∪ … ∪ yr

which is defined by the membership function

μ(y)(y) = max[ 1 ( y), 2 ( y), , r ( y)] for y ∈Y....................(15)

8. Answer:

Application in medicine and economics are as follows:


a. It is used to diagnose the following:
● Dyspnea
● Orthopnea
● Palpitation
● Tachycardia
● Cyanosis
● Oedema
b. Some of application in economics are as follows:
● Economic Modeling
● Industrial Cluster
● Economic Order Quantity

Вам также может понравиться