Beruflich Dokumente
Kultur Dokumente
Question Bank on
Soft Computing
(Neural Network, Fuzzy Logic & Genetic Algorithms)
(Group A Objective Type Question)
1. Bias is used to _________
a. To qualify the threshold value. c. To calculate the net value.
b. To improve the network performance. d. To degrade the network performance
2. The ________ is a single long fiber that carriers the signals from the cell body to other neuron.
a. Axon b. Cell body c. Dendrites d. Lobe
3. The point of contact between an axon of one cell and a dendrite of another cell is known as…
a. Axon b. Cell body c. Synapse d. Soma
4. In sigmoidal function, the graph is _____________
a. V‐Shaped b. S‐Shaped c. Step‐Shaped d. None
5. The layers between input and output layer are ____ layers.
a. Kohonen b. Grossberg c. Hidden d. Vapnik
6. Networks which have feedback connection are called __________.
a. Feedforward b. Recurrent network c. Linear d. Stable
7. In ________ learning input pattern is associated with an output pattern
a. Supervised b. Unsupervised c. Reinforced d. Simulated
8. The algorithm is only provided with a given grade / score as a measured of ANN’s
performance, which is ________ type of learning.
a. Supervised b. Unsupervised c. Reinforced d. Simulated
9. The aim of delta rule is to minimize the _______ of overall training patterns
a. Output b. Error c. Weights d. Bias
10. If two neurons on either side of a synapse are ___________ activated, then the strength of that
synapse is selectively increased.
a. Randomly b. Statistically c. Asynchronously d. Synchronously
11. Boltzmann machine uses _________ neurons with a probabilistic firing mechanism.
a. Deterministic b. Non‐deterministic c. Conditional d. Stochastic
12. In _______ learning all neurons compete in input pattern and the winning neuron undergoes
weight adjustment.
a. Supervised b. Competitive c. Boltzmann d. Reinforcement
13. A 3‐input neuron is trained to output a zero when the input is 110 and a one when the input is
111. After generalisation, the output will be zero when and only when the input is:
(a) 000 or 110 or 011 or 101 (b) 010 or 100 or 110 or 101
(c) 000 or 010 or 110 or 100
14. A perceptron is:
(a) a single layer feed‐forward neural network with preprocessing
(b) an autoassociative neural network
(c) a double layer autoassociative neural network
15. An autoassociative network is:
(a) a neural network that contains no loops
(b) a neural network that contains feedback
(c) a neural network that has only one loop
16. Which of the following is true?
(i) On average, neural networks have higher computational rates than conventional computers.
By - J. P. Singh, Academy of Technology
Page |2
(ii) Neural networks learn by example.
(iii) Neural networks mimic the way the human brain works.
(a) all of them are true (b) (ii) and (iii) are true (c) (i), (ii) and (iii) are true
17. Which of the following is true for neural networks?
(i) The training time depends on the size of the network.
(ii) Neural networks can be simulated on a conventional computer.
(iii)Artificial neurons are identical in operation to biological ones.
(a) all of them are true. (b) (ii) is true. (c) (i) and (ii) are true.
18. What are the advantages of neural networks over conventional computers?
(i) They have the ability to learn by example
(ii) They are more fault tolerant
(iii)They are more suited for real time operation due to their high ʹcomputationalʹ rates
(a) (i) and (ii) are true (b) (i) and (iii) are true (c) all of them are true
19. Which of the following is true?
Single layer associative neural networks do not have the ability to:
(i) perform pattern recognition
(ii) find the parity of a picture
(iii)determine whether two or more shapes in a picture are connented or not
(a) (ii) and (iii) are true (b) (ii) is true (c) all of them are true
20. Learning process according to environment is __________.
a. Competitive learning c. stochastic learning
b. Reinforcement learning d. Adaptation
21. The Hebbian learning rule is _________ type of learning.
a. Supervised b. Competitive c. Boltzmann d. Reinforcement
22. Perception is a ________ type of learning process.
a. Supervised b. Unsupervised c. Reinforced d. Memory based
23. Single layer perception is used for _________ .
a. Linear separability c. Non‐linear separability
b. Error minimization d. Annealing
24. When W(k0) = W(k0 + 1) = W(k0 + 2) = ……….,then we can say perception is _________
a. Linear separable c. Non‐linear separable
b. Converged d. Non‐converged
25. In perception __________ is the formula to change the bias.
a. bnew = wold + exi b. bnew = bold + ewixi c. bnew = bold + e d. bnew = bold + wioi
26. The neuron model in back propagation network includes a ______ type of activation function.
a. Linear b. Non‐linear c. Symmetrical d. Asymmetrical
27. f = (1 / 1 + e‐net k) is known as ______ function.
a. Symmetrical b. Asymmetrical c. Sigmoidal d. Linear
28. X‐OR problem can be solved by __________.
a. Single layer Perceptron c. Multi‐layer perceptron
b. Bayes theorem d. Both (a) and (b)
29. Back propagation is related with _______ type of network.
a. Feed backward b. Feed forward c. Recurrent d. Linear
30. A _____ learning rate in back propagation, results in long training times.
a. Large b. Small c. Medium d. Very large
31. The learning time can be reduced, if the _______ are chosen small at the beginning.
a. Weights b. Inputs c. Learning rate d. Bias
By - J. P. Singh, Academy of Technology
Page |3
32. Wakening or eliminating certain synaptic weights in a selective and orderly fashion is known as
__________ techniques.
a. Cross‐validation b. Generalization c. Network pruning d. Optimization
33. Self‐organizing feature‐maps learn to classify _____________ .
a. Input vectors b. Output vectors c. Weight matrix d. Winner neuron
34. The SOM learning is a completely __________ .
a. Supervised b. Unsupervised c. Reinforced d. Feedback
35. In SOM learning we have to change the weight of ____________ .
a. Only winner c. Not only the winner but also it’s neighbors
b. All d. Only neighbors
36. LVQ learning is __________ type of learning.
a. Supervised b. Unsupervised c. Reinforced d. None
37. In Vector‐quantization, input‐vectors can be divided into areas that are assigned as _________.
a. Voronoi vectors b. Learning vectors c. Vectors d. Code book vectors
38. The Hopfield network consists of a set of neurons forming a multiple loop ________ system
a. Unidirectional b. Parallel c. Feedback d. Feedforward
39. In _______ mode of training, all neurons in Hopfield networks fire at the same time.
a. Stable status b. Output c. Synchronous d. Asynchronous
40. In _______ mode of training, all neurons in Hopfield networks fire at random.
a. Stable status b. Output c. Synchronous d. Asynchronous
41. In associative memory if xk is the input and yk is the output, and when xk ≠ yk, the process is
known as ___________ .
a. Autoassociative b. Nonassociative c. Heteroassociative d. Biassociative
42. The BAM is a __________ network.
a. Autoassociative b. Nonassociative c. Heteroassociative d. None
43. In neural network, the activation function is expressed in terms of
a. Bias b. induced local filed vector c. Input d. none of this
44. In SOM learning we have to change the weight of ____________ .
a) Only winner b) Not only the winner but also it’s neighbors c) All d) Only neighbors
45. Wakening or eliminating certain synaptic weights in a selective and orderly fashion is known as
__________ techniques.
a. Cross‐validation b. Generalization c. Network pruning d. Optimization
46. Back propagation is related with _______ type of network.
a. Feed backward b. Feed forward c. Recurrent d. Linear
47. In Boltzmann machine it is found that units in each row and column are fully interconnected
with the weight of each connection p where
a) p indicates the penalties for violating the condition that almost one unit be “on” in each
row and column
b) p indicates the probability of connectivity of units in each row with the units of each column
c) p indicate the probability of net accepting a change in state for unit xt
d) p be the initialized random weights closed to zero.
48. The neighborhood scheme for SOM is used
a. to find out the neurons of which weight vector to be updated along with winner neuron
b. to find out the neurons of which weight vector to be updated excluding winner neuron
c. to converge the net faster
49. In fuzzy logic controller the rule base is updated by the help of
a) Fuzzy logic b) ANN and GA
By - J. P. Singh, Academy of Technology
Page |4
c) Fuzzy logic, ANN and GA d) none of this
50. Consider two fuzzy sets ‘A’ and ‘B’ which are as follows:
A = {(0, 0.2), (1, 0.5), (3, 0.3)} B = {(0, 0.4), (1, 0.6), (3, 0.9)}. then A‐B will be
a) {(0, 0.2), (1, 0.1), (3, 0.1)} b) {(0, 0.2), (1, 0.5), (3, 0.1)}
c) {(0, 0.2), (0, 0.1), (0, 0.1)} d) None of this.
51. Fuzzy set theory was introduced by………….
a) Zadeh b) Rosenblatt c) Minsky d) Glover
52. A linguistic variable enables its value to be described
a) By a crisp set b) Only qualitatively by a linguistic term
c) Only quantitatively by a corresponding membership function
d) Both quantitatively by a linguistic term and quantitatively by a corresponding membership function.
53. Let A and B are two fuzzy sets with membership function μ, then μA U B(x) is equal to
a) {μ A(x) + μ B(x)} b) { μA(x) ‐ μB(x)} c)min{ μA(x) ,μB(x) } d) MAX { μA(x) ,μB(x) }
54. Let A be a fuzzy set with membership function μ and X be the universal set. Then the support of
the fuzzy set AєX can be defined as
a) S (A) = o+A b) S (A) =IA C) S (A) =Sup {μA(x)}∀ x € X d) S (A) =∑ μA(x)∀ x є X
55. The core of the fuzzy A set is defined by those elements x of the universe such that
a) µA(x) =1 b) µA(x) =0 c) 0 < µA(x) <1 d) 0 ≤ µA(x) ≤1
56. The boundary of the fuzzy A set is defined by those elements x of the universe such that
a) µA(x) =1 b) µA(x) =0 c) 0 < µA(x) <1 d) 0 ≤ µA(x) ≤1
57. Let A and B are two fuzzy sets with membership function μ, then μA ∩ B(x) is equal to
a) {μ A(x) + μ B(x)} b) { μA(x) ‐ μB(x)} c)min{ μA(x) ,μB(x) } d) MAX { μA(x) ,μB(x) }
58. A fuzzy number is a fuzzy set with the property of
a) only normal b) only convex c) both normal and convex d) normal but not convex
59. A normal fuzzy set is one whose one membership function has
a) at least one element x in the universe whose membership value is 1.
b) all elements in the universe have membership value is 1.
c) none of the elements in the universe has membership value of 1
d) at least one element x in the universe whose membership value is 0.
60. The fuzzy set shown in figure 1 is
a) Only normal
b) Only convex
c) both normal and convex
d) convex but not normal
61. The height of a fuzzy set A is the ……………… value of the membership function.
a) Minimum b) Maximum c) mean d) complete
62. The crossover points of a membership function of a fuzzy set is defined as the elements in the
universe of a fuzzy set A for which
a) µA(x) =1 b) µA(x) =0.5 c)µA(x) =0 d) µA(x) =0.3
63. Which of the following(s) is/are found in Genetic Algorithms?
(i) Evolution (ii) selection (iii) reproduction (iv) mutation
(a) i & ii only (b) i, ii & iii only (c) ii, iii & iv only (d) all of the above
64. Matching between terminologies of Genetic Algorithms and Genetics:
Genetic Algorithms Genetics (biology)
(W) representation structures (i) external disturbance, such as cosmic radiation
(X) crossover (ii) chromosomes
By - J. P. Singh, Academy of Technology
Page |5
(Y) mutation (iii) survivability
(Z) selection (iv) sexual reproduction
a) W‐
65. Where are Genetic Algorithms applicable?
(i)real time application (ii) biology (iii) Artificial Life (iv) economics
(a) i, ii & iii only (b) ii, iii & iv only (c) i, iii & iv only (d) all of the above
66. Which of the following(s) is/are the pre‐requisite(s) when Genetic Algorithms are applied to
solve problems?
(i) encoding of solutions (ii) well‐understood search space (iii) method of evaluating the
suitability of the solutions (iv) contain only one optimal solution
(a) i & ii only (b) ii & iii only (c) i & iii only (d) iii & iv only
67. Which of the following statement(s) is/are true?
(i) Genetic Algorithm is a randomised parallel search algorithm, based on the principles of
natural selection, the process of evolution.
(ii) GAs are exhaustive, giving out all the optimal solutions to a given problem.
(iii) GAs are used for solving optimization problems and modeling evolutionary phenomena in
the natural world.
(iv) Despite their utility, GAs remain a poorly understood topic.
(a) i, ii & iii only (b) ii, iii & iv only (c) i, iii & iv only (d) all of the above
68. In GA term ‘Gene’ is termed as
a) Coded design vector b) Coded design variable c) Every bit d) Solution vector
69. A particular set of genes in genome is called
a) Alleles b) chromosome c) Locus d) genotype
70. If crossover between chromosomes in search space does not produce significantly different
offsprings, what does it imply? (if offspring consist of one half of each parent)
(i) The crossover operation is not successful.
(ii) Solution is about to be reached.
(iii) Diversity is so poor that the parents involved in the crossover operation are similar.
(iv) The search space of the problem is not ideal for GAs to operate.
(a) ii, iii & iv only (b) ii & iii only (c) i, iii & iv only (d) all of the above
71. The size of each chromosome for the problem of maximizing the function f(x)=x2 in the interval 0
≤ x ≤ 31 is
a. Six b. five c. Four d. three
(Group B 5 Marks Questions)
1. What is meant by supervised, reinforcement and unsupervised learning rules?
2. Define soft ‘computing’. How does it differ from traditional hard computing? What are the basic
features of an ANN?
3. Discuss the fixed increment perception learning algorithm for a classification problem with n
input features (x1, x2, x3………….. xn) and two output classes (0/1).
4. Compare and contrast biological neurons and artificial neurons.
5. Described the discrete Hopfield Net and its training algorithm.
6. What is delta learning rule? Why is it called as Least Mean square rule?
7. Implement the AND function using McCulloh‐Pitts neurons using binary data.
8. Compare feed‐forward and feedback network. In what ways is bipolar representation better
than binary representation?
9. Using Hebb’s rule, find the weights required to perform following classifications. The vectors (1,
‐1, 1, ‐1) and (1, 1, 1, ‐1) belong to class 1 (target value 1) and (‐1, ‐1, 1, 1) and (1, 1, ‐1, ‐1) belong
to class 2 (target value ‐1). Test the response of the network using each training set.
10. What is the neccessity of activation function in artificial neural network? What is a learning rate
parameter?
11. Implement OR function with binary inputs and bipolar targets using Perceptron training
algorithm upto 3 epochs.
12. With a neat flowchart, explain the training process of Perceptron network.
13. Define overfitting or over training.
14. Explain the function of multi‐layer perception (MLP).How does it differ from a single layer
perception (SLP)?
15. Define ANN. What is the role of an activation function in an ANN model? Name two activation
functions.
16. a) Consider the linguistic variable “Age”. Let the term ‘old’ be defined by
⎧0 if x € [0,40]
µ old = ⎨ - 2 -1
⎩(1 + ((x - 40) / 5) ) if x € [40,100]
Determine the membership function of the terms “very old”,”not very old” and ”more or less
old”.
b) State the different types of decomposition theorem of fuzzy logic.
17. a) Let A be a Fuzz set defined by: A = 0.5/x1 + 0.4/ x2 + 0.7/x3 + 0.8/x4 + 1/x5. List all α‐cuts
and strong α‐cuts of A.
b) State the extension principle in fuzzy logic.
18. a) Define the T‐norm with example.
b) Define T‐co norm with example.
c) Explain aggregation with an example. Also mention the necessary condition to be satisfied by
an aggregation operation.
19. Let A be fuzzy set defined as
A=0.5/x1+0.4/x2+0.7/x3+0.8/x5+1.0/x8
Enlist all α‐cuts and strong α‐cuts of A.
20. let A = “ near 0.74” = { 0/0.730 + 0.8/0.735 + 1/0.740+ 0.3 / 0.750} and
B = more or less 0.74” = { 0.2/0.730 + 0.6/0.735 + 0.8/0.740 + 1/ 0.745 + 0.5 / 0.750}
For the two sets find a) Α ∪ Β b) Α ∩ Β c) ¬ Α
21. Let A and B be the two fuzzy numbers whose membership function is given bellow:
⎧ (x + 2) ⎧ (x - 2)
⎪ 2 :-2 < x ≤ 0 ⎪ 2 :2 < x ≤ 4
⎪ ⎪
⎪ (2 - x) ⎪ (6 - x)
A(x) = ⎨ :0 < x < 2 B (x) = ⎨ : 4 < x < 6
⎪ 2 ⎪ 2
⎪0 : otherwise ⎪0 : otherwise
⎪ ⎪
⎩ ⎩
Calculate the fuzzy numbers A+ B, B + A, A – B and B – A.
22. In a class of 10 students (the universal set), 3 students speaks German to some degree, namely
Alice to degree 0.7, Bob to degree 1.0, Cathrine to degree 0.4.
a) What is the size of the subset A of German speaking students in the class?
b) What is the support of the fuzzy set A?
c) What is 0.5A (the 0.5‐cut of the fuzzy set A) ?
d) What is 0.4+A (the strong 0.4‐cut of the fuzzy set A) ?
e) Is A normal?
By - J. P. Singh, Academy of Technology
Page |7
23. Explain different types of selection methods in Genetic algorithms.
(Group C 15 Marks Questions)
24. a) What are artificial neural networks? Show a model of an artificial neuron.
b) Describe the neural network architecture.
c) What do you mean by supervised, reinforcement and unsupervised learning rules?
25. a) Outline the basic structure and components of a simple biological neuron.
(b) Describe how this is related to a McCulloch‐Pitts neuron.
(c) Design networks of McCulloch‐Pitts neurons that implement logical NOT gate. Draw the
network and label all the weight and threshold values. [5+5+5]
26. a. Explain a single layer Perceptron network with suitable diagram. [2+3+4+4+2]
b. State the Perceptron learning rule and apply that rule to find a Perceptron network for AND
function with bipolar inputs and targets. [Take initial weights including bias and learning
rate to be 0]. Find out the final weights and bias.
c. Derive the weight update formula in Back Propagation rule for multi layer feed forward
network using differentiable activation function. What are merits and demerits of Back
Propagation Algorithm?
27. a) Describe the architecture of Hopfield networks.
b) Write down the algorithm to train the Hebb net.
c) State the delta learning rule.
28. a. Describe the architecture of Kohonen self organizing feature maps. [5+5+5]
b. State the training algorithm for self organizing maps.
c. Consider a Kohonen net with two cluster units and 3 input units. The weight vectors for the
cluster unit are [0.9, 0.7, 0.6] and [0.4, 0.3, 0.5]. Find the wining cluster unit for the input vector
[0.4, 0.2, 0.1]. Find the new weight for the winning unit using a learning rate of 0.2.
29. An athletic race was conducted. The following membership functions are defined based on the
speed of the athletes.
⎧ 1 0.1 0.3 ⎫
Low= ⎨ + + ⎬
⎩100 200 300 ⎭
⎧ 0.5 0.57 0.6 ⎫
Medium = ⎨ + + ⎬
⎩100 200 300 ⎭
⎧ 0.8 0.9 1 ⎫
High= ⎨ + + ⎬
⎩100 200 300 ⎭
Find the following
R=Low X Medium
S= Medium X High
T=R 0 S using max‐min composition
T=R 0 S using max‐product composition
30. Let A and B be the two fuzzy numbers whose membership function is given bellow:
⎧( x + 2) / 2 for - 2 < x ≤ 0
⎪
A( x) = ⎨(2 - x) / 2 for 0 < x < 2
⎪0
⎩ otherwise
⎧( x - 2) / 2 for 2 < x ≤ 4
⎪
B ( x) = ⎨(6 - x) / 2 for 4 < x < 6
⎪0
⎩ otherwise
Calculate the fuzzy numbers A+B, A ‐B, B ‐A, A.B, A/B, MIN(A, B), MAX(A, B).
31. a) Consider the following age groups: 0‐10; 10‐20; 20‐30; 30‐40; 40‐50; 50‐60; 60‐70; 70 and above.
The fuzzy sets “young”, ”middle” and “old” are represented by the membership function graphs as
illustrate in fig. given below. Find the union, intersection, complement and product operation with
the help of membership function.
32. a. Consider a fuzzy set A in the universe whose membership function is given by
µA(x) = 1 ‐ |x‐2| for |x‐2| ≤ 2 [5 + 10]
= 0 otherwise
i. Sketch the membership function
ii. What is the support set of A?
iii. What is the α cut of A for α = 0.5
b. Consider the knowledge base represented by the set of fuzzy rules Y ‐> U with
Y=0.2/y1+0.6/y2+1.0/y3+0.6/y4+0.2/y5
U=0.3/u1+1.0/u2+0.3/u3
Determine the membership function matrix R of this knowledge base.
Determine the respective fuzzy inference U’ for three fuzzy observations Y’ whose
membership functions are
µY’(yi)=[0.4 1.0 0.4 0 0]
µY’(yi)=[0.4 0.8 0.4 0 0]
µY’(yi)=[0.4 0.5 0.4 0 0]
33. Given the following fuzzy number and using Zadeh’s extension principle, Calculate K= I * J and
explain why 6 is non‐convex.
I = approximately 3 = { 0.2/2 + 1/3 + 0.1/4}
J = approximately 2 = { 0.1/1 + ½ + 0.3/3}
34. For fluids, the product of the pressure (p) and volume V of the fluid is a constant for a given
temperature i.e. PV = constant
Assume that at a given temperature a fluid of fuzzy volume
V1 = {0.0/0.5 + 0.5/0.75 + 1.0/1.0 + 0.5/1.25 + 0.0/1.5} is under a fuzzy pressure
P1 = {0.0/0.5 + 0.5/1.75 + 1.0/2.0 + 0.5/2.25 + 0.0/2.5}
a) Using the extension principle, determine the pressure P2, if the volume is reduced to
V2 = { 0.0/0.4 + 0.5/0.45 + 1.0/0.5 + 0.5/0.55 + 0.0/0.6}
b ) Develop analogous continuous membership function for the fuzzy pressure P1 and volume
V1 and solve for the pressure P2 using (i) the vertex method and (ii) the DSW algorithm. Plot
the resulting membership function.
c) Explain why P2.V2 would not be same as P1.V1.
35. In the field of computer networking there is an imprecise relationship between the level of use
of a network communication bandwidth and the latency experienced in peer –to peer
communications. Let X be a fuzzy set of use levels and Y be a fuzzy set of latencies with the
following membership functions.
By - J. P. Singh, Academy of Technology
Page |9
X={0.2/10 + 0.5/20 + 0.8/40 + 1/60 + 0.6/80 + 0.1/100}
Y={0.3/0.5 + 0.6/1 + 0.9/1.5 + 1.0/4 + 0.6/8 + 0.3/20}
a) Find the Cartesian product represented by the relation R = X + Y
Now suppose we have a second fuzzy set of bandwidth usage given by
Z={0.3/10 + 0.6/20 + 0.7/40 + 0.9/60 +1/80 + 0.5/100}
Find S = Z * R
b) using max‐min composition
c) using max‐product composition
36. Consider the following two discrete fuzzy sets, which are defined on universe X={‐5, 5}
A= Zero ={ 0/‐1 + 0.5/‐1 + 1/0 + 0.5/1 +0/2}
B= postive medium = { 0/0 + 0.5/1 + ½ +0.5/2 + 0/4}
a) Construct the relation for the rule IF A THEN B using the Mamdani Implication and the
product implication.
If we introduce a new antecedent,
Aʹ= positive small= {0/‐1 + 0.5/0 + 1/1 + 0.5/2 + 0/3}
b) Find the new consequent Bʹ, using max‐min composition
37. a) What do you mean by breeding? Discuss the various steps of breeding cycle. [1+3+4+3+4]
b. Give examples to illustrate various cross over techniques in GA.
c. Differentiate between the single point and multi point mutation in GA.
d. Explain the various stopping condition used in genetic algorithm.
38. a) How is genetic algorithm different from traditional algorithm?
b) State the schema theorem.
c) Discuss the different types of crossover method in Genetic algorithm.
d) Differentiate between the single point and multi point mutation in GA.
e) Explain any one selection strategy used in GA.
39. a) State and prove the schema theorem on Genetic algorithm.
b) A population contains the following strings and fitness values at generation 0:
# string Fitness
1 10001 20
2 11100 10
3 00011 5
4 01110 15
The probability of mutation is pm=0.01 and the probability of cross‐over is Pc=1.0. Calculate
the expected number of schemata of the form 1**** in Generation 1. Estimate the expected
number of schemata of the form 0**1* in generation 1.
40. a) Explain the cycle of Genetic algorithm.
b) What are the cross‐over rate and mutation rate?
c) Solve the following traveling salesman problem for 6 cities by using genetic algorithm.
City 1 2 3 4
1 0
2 5 0
3 6 13 0 0
4 8 8 9 4
41. Write short notes on any three of the following: [5+5+5]
By - J. P. Singh, Academy of Technology
P a g e | 10
b. Tabu search.
c. Simulated annealing.
d. Hopfield Net.
e. Different defuzzification operators.
f. Ant Colony Optimization