Qno.

Question Option1 Option2 Option 3 Option 4 Correct option

1 Soft computing deals with imprecision probabilities both A and B None of above C

2 Which of the following is true Genetic Algorithms Genetic Algorithms both A and B None of above A
about Genetic Algorithms work with the work with the parameter
coding of the themselves
parameter set.
3 Which of the following data linguistic numeric both A and B None of above C
can be handle by Neuro fuzzy
hybrid system
4 Which of the following is It does not need neuro fuzzy hybrid can both A and B None of above C
correct about neuro fuzzy previous knowledge mimic human decision-
logic of relationships of making process
data
5 Which of the following is Soft computing Soft computing deals both A and B None of above B
correct about soft computing deals with only with approximate
precise models models

6 For which of the following Classification Pattern Recognition both A and B None of above C
purpose we can use Artificial
neural network
 7 Which of the following is/are supervised unsupervised both A and B None of above C
type of learning

8 Which of the following is It deal with fuzzy It is based on is based on both A and B None of above C
correct about soft computing logic empirical data

9 Which of the following is Conventional Conventional Artificial both A and B None of above C
correct about Conventional Artificial Intelligence methods are
Artificial Intelligence Intelligence deal limited by symbols
with prdicate logic
10 Which of the following is Fuzzy set theory Fuzzy logic is usually both A and B None of above C
correct defines fuzzy represented as IF-THEN
operators on fuzzy rules
sets.

11 Which of the following is true algorithm which can precise solutions both A and B None of above A
about soft‐computing easily adapt with the
methodology change of dynamic
environment
12 Fundamental unit of network Brain Neuron Nucleus Axon B
is
13 Which of the following is true (ii) is true (i) and (ii) are true (i) and (iii) are true All of the mentioned C
for neural networks?
(i) The training time depends
on the size of the network.
(ii)Artificial neurons are
identical in operation to
biological ones. (iii)
Neural networks can be
simulated on a conventional
computer.
14 Which of the following gives Stochastic gradient Convoultion operator Sigmoid activation Non-zero bias C
non-linearity to neural decent
network
15 Applications of Neural Pattern Recognition clustering Classification All of the above D
Network includes
16 What is the advantage of basis Storing in basis Training of basis Training of basis function is None of the mentioned B
function over MLFFNN ? function is faster function is faster than slower than MLFFNN
than MLFFNN MLFFNN
17 Which is true for neural Each node computes Node could be in excited It has set of nodes and All of the mentioned D
networks? it’s weighted input state or non-excited state connections

18 Which of the following (i) and (ii) are true (ii) is true (i) and (iii) are true (i) is true D
statement true in case of
Radial Basis Function
networks?
(i) Updating of connecting
weights does not occur
between inout nodes and
hidden layer.
(ii)The number of hodden
layers is kept more than 2
(iii) Updating of connecting
weights does not occur
between hidden layer and
output layer.
(iv)It is a three-layered
network
19 An artificial neuron receives n Σ wi Σ wi + Σ xi Σ wi* xi Σxi C
inputs x1, x2, x3............xn
with weights w1, w2,
..........wn attached to the input
links. The weighted
sum_________________ is
computed to be passed on to a
non-linear filter Φ called
activation function to release
the output.
20 What are dendrites? nuclear projections fibers of nerves other name for nucleus None of the mentioned B

21 Identify the following Sigmoid function Ramp function Step function Gaussian function A
activation function :
φ(V) = Z + (1/ 1 + exp (– x *
V + Y) ), Z, X, Y are
parameters
22 A perceptron can correctly Overlapping Non-Linearly seprable Linearly seprable None of the mentioned C
classify instances into two
classes where the classes are
23 In a feed- forward networks, Unidirectional Bidirectional Multidirectional Directiona A
the connections between
layers are ___________ from
input to output.
24 Each connection link in ANN neurons bias activation function weights D
is associated with
________ which has
information about the input
signal.
25 What is true regarding There is no feedback It is also called Error in output is propagated All of the above D
backpropagation rule? of signal at nay generalized delta rule backwards only to determine
stage weight updates

26 Layers between the input and Resultant layer Hidden layer Multilayer None of the above B
output layers are known as:
27 What is the limitation of No limitation it can It is a weak algorithm It can only solve problems which It cannot converge to a C
perceptron learning rule? solve almost any are linearly seperable correct solution,even if there
problem exists a solution

28 Can Ex-OR gate be Yes No Maybe Cannot be determined B

implemented using single
layer perceptron ?
29 How can false minima be by providing by using probabilistic can be either probabilistic update none of the mentioned B
reduced in case of error in additional units update or using additional units
 recall in feedback neural
networks?

30 Function of dendrites is? receptors transmitter both receptor & transmitter none of the mentioned A

31 What is the sequence of steps 1,4,3,2 2,1,4,3 1,2,3,4 2,3,4,1 B

followed in training a
perceptron?
1. For a sample input,
compute an output
2. Initialize weights of
perceptron randomly
3. Go to the next batch of
dataset
4. If the prediction does not
match the output, change the
weights
32 What is perceptron in Neural It is double layer It is an auto-associative It is a single layer feed-forward It is a neural network that C
Network auto-associative neural network neural network with pre- contain feedback
neural network processing

33 A 4-input neuron has weights 76 238 119 123 B

1, 2, 3 and 4. The transfer
function is linear with the
constant of proportionality
being equal to 2. The inputs
are 4, 10, 5 and 20
respectively. The output will
be:
34 Which application out of wall following gesture control rotating arm and legs wall climbing A
these of robots can be made of
single layer feedforward
network?
35 ----------------is a store house Associative memory Commutative memory Neural networks Memory A
of associated patterns which
are encoded in some form
36 If the associated pattern pairs Auto associative Hetero associative neuro associative memory none B
(x,y) are different and if the memory memory
model recalls a y given an x
or vice versa, then it is termed
as -------------
37 What type of inputs does ART bipolar binary both bipolar and binary none of above B
– 1 receives?
38 What type of learning used in Competitive Hebbian Gradient Descent none of above A
comparison field in ART?
39 What should be the aim of to capture inputs to feedback the captured to capture the behaviour of none of above D
training procedure in outputs system
boltzman machine of
feedback networks?
40 What is synchronous update all units are updated a unit is selected at a predefined unit is selected and none of above A
in hopfield model? simultaneously random and its new state its new state is computed
is computed
41 What is hebbian learning? synaptic strength is synaptic strength is synaptic strength is proportional none of above A
proportional to proportional to to correlation between firing of
correlation between correlation between presynaptic neuron only
firing of post & firing of postsynaptic
presynaptic neuron neuron only
42 What consist of competitive feedforward paths feedback paths either feedforward or feedback combination of feedforward D
learning neural networks? and feedback
43 What consist of boltzman fully connected asynchronous operation stochastic update All of the above D
machine? network with both
hidden and visible
units
44 Self organizing feature map Kohonen John holland Stephen grossberg none of above A
neural net was developed by
45 In Self Organizing Feature Competion Co-operation Adaptation none of above C
map weight of winning
neuron and surrounding to
 winning neuron is updated in
which step?

46 In self organizing feature map Neighborhood activation function sigmoidal funtion All of the above A
neighbor of winning neuron is function
determined by
47 If target output is not Unsupervised Supervised Hebbian Stochastic A
available and system learn of
its own by discovering and
adapting to structural features
in the input patterns.This type
of training is called....
48 How are weights updated in updated for winning updated for neighbours updated for winning unit and its none of above C
Self Organizing Feature Map? unit only of winner only neighbours
49 Hebbian Learning Rule was Stephen Grossberg Donald Heb KOHONEN Charles Darwin B
proposed by.......
50 How are input layer units feedforward manner feedback manner feedforward and feedback feedforward or feedback A
connected to second layer in
competitive learning
networks?
51 Fuzziness means ----------- Vagueness Clear Precise Certainty A

52 ----------------- truth values are crisp logic boolean logic fuzzy logic none C
multivalued.
53 Fuzzy logic propositions are fuzzy fuzzy qualifiers fuzzy quantifiers none C
also quantified by --------------
54 Fuzzy inference also referred approximate reasoning fixed reasoning none A
to as -------------- reasoning
55 Conversion of a fuzzy set to fuzzification defuzzification fuzzy logic fuzzy rule A
single crisp value is called ----
-------------
56 .______________ is/are the fuzzy logic probability entropy all of these D
way/s to represent
uncertainty.
57 Fuzzy Set theory defines Union intersection complement all of these D
fuzzy operators. Choose the
fuzzy operators from the
following
58 A __________ point of a core Support crossover none of above C
fuzzy set A is a point x ∈ X at
which µA(x) = 0.5
59 Both fuzzy logic and artificial Both gives precise Artificial neural network In each, no precise mathematical Fuzzy gives exact result but C
neural network are soft and accurate results. gives accurate result, but model of the problem is required. artificial neural network does
computing techniques because fuzzy logic does not. not.
60 Fuzzy Genteic hybrid system Fuzzy logic in Fuzzy logic controlled Genetic algorithm controlled None of above B
is a parallel with genetic genetic algorithm fuzzy logic
algorithm
61 Fuzzy logic is a form of Two-valued logic Crisp set logic Many-valued logic Binary set logic C

62 Fuzzy relation is a fuzzy set single set crisp set intersection set union set B
defined on the Cartesian
product of -----------.
63 How is Fuzzy logic different If and then approach for approach do approach while approach A
from conventional control
methods?
64 The room temperature is hot. Fuzzy Set Crisp Set both of these None of above A
Here the hot (use of linguistic
variable is used. can be
represented by
65 What are the following Fuzzification → Fuzzification → Rule evaluation → Fuzzification Rule evaluation → A
sequence of steps taken in Rule evaluation → Defuzzification → Rule → Defuzzification Defuzzification →
designing a fuzzy logic Defuzzification evaluation Fuzzification
machine ?
66 Matrix crossover is also one dimensional Two dimensional Three dimensional none B
known as------------------
67 ---------- of bit involves Mutataion Crossover Inversion Segregation A
changing bits from 0 to 1 and
1 to 0
68 --------------is a process in Inversion Conversion masking Segregation C
which a given bit pattern is
transformed into another bit
pattern by means of logical
bit-wise opearion.
69 ------------ performs linear Linear+end Discrete inversion Continuous inversion Mass inversion A
inversion with a specified inversion
probaility of 0.75
70 Possible settings of traits are locus alleles genome genotype B
called in genes----------
71 ----------means that the Recombination Selection Mutataion none C
element of DNA is modified.
72 ------------mmic the principle Genetic Genetic algorithm Genetic Evolution None B
of natural genetics programming
73 ------------mmicis the swarm intelligence Ant colony Genetic Algorithm none A
behaviour of social insects.
74 The-------------- of an Strength fitness Gene Chromosome B
organisim is measured by
means of success of organisim
in life.
75 -------------------is a way of Convesion encoding coding none B
representing individual genes.
76 Population size,Mutation rate control paremeter central parameter connection parameter none A
and cross over rate are
together referred to as
77 The space for all possible space search search space area C
feasible solutions is called
78 --------------------is the first Reproduction Recombination Mutataion none A
operator applied on
population.
79 Crossover operator proceeds 4 3 5 2 B
in -------- steps
80 Reproduction operator is also Recombination Selection Regeneration none B
known as --------------
81 Genetic algorithms belong Artificial Optimization area Complete enumeration familty Non computer based human A
to the family of methods in Intelligence area methods solutions
the
82 Which GA operation is Initial population selection of sub- Reproduction to produce next convergence testing C
computationally expensive creation population for mating generation

83 Which GA encoding scheme Binary coded GA Real coded GA order GA Tree encoded GA A
gives faster execution?
84 The basic idea behind Ant of problem solvers of candidate solutions to of candidate solutions in which of problem solvers each of A
Colony Optimization that interact with try and create better each candidate is heuristically which does an independent
algorithms is to work with a each other through candidates by mixing refined. heuristic search.
population signs genes.

85 A genetic algorithm (GA) for a small population a large population of fit a small diverse population of fit a large diverse population of D
optimization is most likely to of fit and similar and similar individuals. individuals. fit individuals.
succeed given individuals.
86 ∆wkj(n) = ηYk(n)xj(n) is the Error correction Boltzmann Learning Hebbian Hypothesis Competitive learning C
formula according to the learning

87 Probability of error in recall patterns are stored inputs are captured weights are chosen appropriately none of the mentioned C
of stored patterns can be appropriately appropriately
reduced if?
88 What should be the aim of Capture the Pattern to feedback the captured to capture the behaviour of Capture the inputs A
training procedure in environment outputs system
boltzman machine of
feedback networks?
89 The components of the fully connected Asynchronus operations Stochastic update All of the above D
boltzman machine are ? network with both
hidden and visible
units
90 Boltzman machine reduces no such method simulated annealing hopfield reduction none of the mentioned B
effect of additional stable exist
states?
 91 For which other task can Pattern reconization Feture classification Pattern Association Pattern mapppin D
boltzman machine be used?
92 Kohonen's SOM is also called Maxnet Self Organinzing Map Topology preserving Non of the above C
map
93 The Kohonen map performs a Continious input Continious input space Descrite input space to descrite Descrite input space to A
mapping from a space to descrite to continious output output space continious output sapce
output space space
94 synaptic weights between Hebb Rule Maxnet function Mexican hat function Bipolar function C
neurons in the Kohonen layer
is achieved by
95 The main process flow which Intila Competition ,Weight Intial weight No such process flow is A
defines the SOM functioning weights,competition, change, corealtion, ,competition,corelation,adaptation required
are coopration, adaptation
adaptation
96 When one cell repeatedly 1963 1949 1950 1945 B
assists in firing another, the
axon of the first cell develops
synaptic knobs (or enlarges
them if they already exist) in
contact with the soma of the
second cell " is Hebb's rule
derived in the year
97 A Hopfield Network is a capacity of netwrok storage capacity of processing power of nework Allof the above B
model of Associtaive Memory netwrok
,which is used to determine
the
98 Adaptive resonance theory is Stephen Grossberg Hebbs , in 1949 Stephen Grossberg and Gail Frank Rosenblatt ,in 1958 A
a type of neural network and Gail Carpenter Carpenter 1985
technique developed in1987
99 In Hamming net --------- is Maxican Net Hamming Net MaxNet None C
used as a subnet
100 ART Consist of -------Unit 2 3 4 1 A

101 ART1 network can be Differntial equation Analog circuit Analog Circuit Governing the None of the above C
implemeted by Differntial equation
102 SOM can be used for the genes in the medical the study of multi- and web based contents All of the Above D
clustering of , field media
103 Constrcut the MaxNet with a1 a4 a2 a3 B
four neuros and inhibitory
weight E=.25, when given the
intial activation signals (input
signals ) . The Intial activatin
set are a1(0)=0.1,
a2(0)=0.3,a3(0)=0.4,a4(0)=0.7
. The Maxnet nuron selected
is :
104 How many output Fuzzy 5 3 4 2 D
Logic produce?
105 Fuzzy Logic is a ? Hardware Software both A and B None of them C

106 Fuzzy Logig Set is ------ of subset superset Multiset neither of them B
Crisp set/Binary logic set
107 Each element of X is mapped Degree of Membership value both A and B None of them C
to a value between 0 and 1. It Membership
is called _____.
108 What action to take when IF Heat Cool Not sure No Change B
(temperature=Warm) AND
(target=cool) THEN?
109 What action to take when IF Heat Cool No Change None of them A
(temperature=Cool) AND
(target=Warm) THEN?
110 Fuzzy Logic is Introduced by Lofti A.Zadhe Hecht Nielsm Kohenen Rosenblatt A
----
111 Consider a fuzzy set A {0, 1, 2, 3, 4, 5, 6, 7, {1, 2, 3, 4, 5, 6, 7, 8, 9, {2, 3, 4, 5, 6, 7, 8, 9, 10} {2, 3, 4, 5, 6, 7, 8, 9, 10,11} B
defined on the interval X = [0, 8, 9, 10} 10}
10] of integers by the
membership Junction
μA(x) = x / (x+2) Then the α
cut corresponding to α = 0.5
will be
112 Union of following two fuzzy {1/2+ 0.4/4 {0.5/2+0.3/4 Onaly A None of them A
set is ,Fuzzyset A={ 1/2+ +0.5/6+1/8 +0.1/6+0.2/8}
0.3/4 +0.5/6 +0.2/8} ,Fuzzy
Set B ={0.5/2 + 0.4/4
+0.1/6+1/8}
113 Fuzzy logic is usually IF-THEN-ELSE IF-ElSE rules NOT ELSE Rules All of the above B
represented as rules
114 Intersetion of following two {1/2+ 0.4/4 {0.5/2+0.3/4 {1.5/2+0.7/4 +0.6/6+1.2/8} All of the above B
fuzzy set is, Fuzzyset A={ +0.5/6+1/8 +0.1/6+0.2/8}
1/2+ 0.3/4 +0.5/6 +0.2/8}
,Fuzzy Set B ={0.5/2 + 0.4/4
+0.1/6+1/8}
115 Complement of follwing {-1/2+-0.3/4+- {0/2+0.7/4+0.5/6+0.8/8} Only B both A and B B
Fuzzy set is :: A={ 1/2+ 0.3/4 0.5/6+ -1/8}
+0.5/6 +0.2/8}
116 Complement of follwing {-0.5/2+ -0.4/4+ - { {0.5+0.6+0.9+0} None of them B
Fuzzy Set B ={0.5/2 + 0.4/4 0.1/6 + -0.2/8} 0.5/2+0.6/4+0.9/6+0/8}
+0.1/6+1/8}
117 ----------are algorithms that Fuzzy Relational fuzzy set crispe set logic None of them B
learn from their more DB
complex environments (hence
eco) to generalize,
approximate and simplify
solution logic.
118 The Inference method is Modular Features angular features Only B Cant say anything B
based on the geometrical
shapes and geometry, whereas
the angular Fuzzy set is based
on the angular features .
119 Neural network can be used Mutivlaued input Fuzzy membership Genetic input values Both b and c B
to obtain Fuzzy membeship values
vlaues
120 Genetic algorithm can be used Neural network Fuzzy membership Cant say anything Only B C
to obtain the Fuzzy based values values
membership values
121 Follwoign Soft comptuing Genetic Algortihm Neural network Evoultionay Theory both A and B D
constituents can be used to
obtain the fuzzy membership
values.
122 The main components of Fuzzification and Knowledge Base Both a and B Only A C
Fuzzy Control systems are Defuzzifiction
123 In Membership function graph Universe of degrees of membership degrees of discourse None of the above A
x-axis represent Discourse in the [0, 1] interval
124 Which of the following is not Automotive Systems Domestic Goods Domestic Control Environment Control C
Application Areas of Fuzzy
Logic?
125 Fuzzificiton is the method of Binary qunatity Multivalued quantity fuzzy quantity fuzzy as well as binary C
transforming a crisp quantity quantity
into a ---
126 Defuuzificaiton is the method Crisp qunatity Multivalued quantity fuzzy quantity fuzzy as well as binary A
of transforming a fuzzy quantity
quantity into a ----
127 Maximum membership Defuzzification Fuzzification both A and B None of the above A
principle, centroid method,
weighted average method,
center of sums, etc are the
method used for---
128 Intuition, inference, rank Defuzzificaiton Fuzzification Only A None of the above B
ordering, angular fuzzy sets,
neural network, etc., are the
methods used for ---
129 Fuzzificaiton use ----------- IF-THEN IF- THEN- ElSE both A and B None of the above A
rules for fuzzifying the crisp
value.
130 Defuzzification uses the centroid of the sets. centroid of the relations Disjoint elements in the sets. None of the above A
center of gravity methods to
find the -----
131 How is Fuzzy Logic different IF and THEN FOR Approach WHILE Approach DO Approach A
from conventional control Approach
methods?
132 The support of Fuzzy set A is µA(x) > 0 µA(x) = 1 µA(x) = 0.5 µA(x) ≠ 1 A
the set of all points x in X (is
the universe of discourse)
such that
133 Fuzzy Expert Fuzzy Modelling Fuzzy Logic Controller All of the above D
What Is Another Name For System
Fuzzy Inference Systems?
134 Mamdani's Fuzzy Inference Control any two Control a steam engine and a C
Method Was Designed To combinations of any Control a television and boiler combination by Control a air craft and feul
Attempt What? two products by remote combination by synthesising a set of linguistic level combination by
synthesising a set of synthesising a set of control rules obtained from synthesising a set of
linguistic control linguistic control rules experienced human operations. linguistic control rules
rules obtained from obtained from obtained from experienced
experienced human experienced human human operations.
operations. operations.

135 Ggather all the different fuzzy Aggregation Fuzzification Defuzzification Normalization A
set outputs and combine them
into a single fuzzy set outputs
is the proecess called as …
136 Genetic Algorithm - Neural locate a locate a neighborhood Cant say anything Option 1 is true B
Network or GANN , has the neighborhood of the of the optimal solution
ability optimal solution quickly over the
quickly traditional serach
strategy
137 The disadvantage of BPN is they can recognize they have the ability to It requires no prior knowledge None C
patterns similar to recognize new patterns; about the weak learner and so can
those they have be flexible.
learnt
138 Simplified Fuzzy ARTMAP is Supervised Learning Unsupervised Learning Reinforcement Learning Both 1 and 2 A
based on
139 The advantages of It performs It helps to find out It ensembles a powerful model All of the Above D
Neuro·Genetic Hybrids optimization of complex structure of that could significantly improve
neural network ANN for given input th
parameters with and the output data set
simplicity, ease of by using its learning rule
operation, minimal as a fitness function.
requirements and
global perspective.
140 Neuro Fuzzy Hybrid sytems Cooperative NFS General NFHS both A and B None of the above C
(NFHS) further can be
classified as
141 Properties of Genetic Neruro Large Population Locate the both a and b None of the above C
Hybrid system are genration heighbourhood of the
optimal solution quickly
142 Perceptron can learn AND XOR both A and B None of these A

143 A perceptron is a --------------- Feed-forward neural Back-tracking algorithm Feed Forward-backward Recurrent Algorithm A
-----------------. network algorithm

144 Identify the following Step function Ramp function Sigmoid function Gaussian function C
activation function :φ(V) = Z
+ (1/ 1 + exp (– x * V + Y)
),Z, X, Y are parameters
145 An artificial neuron receives n Σ wi Σ xi Σ wi + Σ xi Σ wi* xi D
inputs x1, x2, x3............xn
with weights w1, w2,
..........wn attached to the input
links. The weighted
sum_________________ is
computed to be passed on to a
non-linear filter Φ called
activation function to release
the output.
146 Back propagation is a learning Forward from Backward from sink to Forward from source to hidden Backward from sink to B
technique that adjusts weights source to sink source nodes hidden nodes
in the neural network by
propagating weight changes.
147 A neuron with 3 inputs has 0.2 1.00 0.02 -1.00 C
the weight vector [0.2 -0.1
0.1]^T and a bias θ = 0. If the
input vector is X = [0.2 0.4
0.2]^T then the total input to
the neuron is:
148 Which of the following neural (A) only (B) only (A) and (B) only (A) and (C) only A
networks uses supervised
learning?(A) Multilayer
perceptron(B) Self organizing
feature map(C) Hopfield
network
149 Which of the following is All of the mentioned (ii) is true (i) and (ii) are true None of the above mentioned A
true for neural networks I) are true
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
150 Why is the XOR problem Because it can be Because it is a complex Because it can be solved by Because it is simplest D
exceptionally interesting to expressed in a way binary operation that can single layer perceptron linearly inseperable pattern
neural network researchers that allows you to not be solved using that exists
use a neural network neural network
151 In Delta Rule for error weights are adjusted weights are adjusted weights are adjusted w.r.to none of the above B
minimization w.r.to change in the w.r.to difference difference between input and
output between desired output output
and actual output
152 Which is true for neural It has set of nodes Each node computes it’s Node could be in excited state or All of the mentioned D
networks? and connections weighted input non-excited state

153 What is back propagation? It is another name It is the transmission of It is the transmission of error None of the mentioned C
given to the curvy error back through the back through the network to
function in the network to adjust the allow weights to be adjusted so
perceptron inputs that the network can learn

154 Which of the following is not It can explain result It can survive the failure It has inherent parallelism It can handle noise A
the promise of artificial neural of some nodes
network?
155 A perceptron adds up all the TRUE FALSE Sometimes – it can also output Can’t say A
weighted inputs it receives, intermediate values as well
and if it exceeds a certain
value, it outputs a A
perceptron adds up all the
weighted inputs it receives,
and if it exceeds a certain
value, it outputs a
1, otherwise it just outputs a
0.
156 To generate the final output, Smash function sum function Activation function Output function C
the sum is passed on to a non-
linear filter φ called
157 --------------function is a Activation function Thresholding function Signum function Sigmoidal function D
continuous function that
varies gradually between the
asymptotic values 0 and 1 or -
1 and +1
158 ------------------- carrying the Feed forward Fast forward network Fast network Forward network A
weights connect every input network
neuron to the output neuron
but not vice-versa
159 In the learning method, the Supervised learning Unsupervised learning Reinforced learning Hebbian learning B
target output is not presented
to the network ----------------
160 In Supervised Learning is classes are not classes are predefined classes are not required classification is not done B
predefined
161 In which ANN, loops are FeedForward ANN FeedBack ANN both A and B None of the Above B
allowed
162 The output at each node is node value Weight neurons axons A
called_____.
163 In FeedForward ANN, unidirectional bidirectional multidirectional All of the above A
information flow is
_________
164 Which of the following is not Unsupervised Reinforcement Learning Supreme Learning Supervised Learning C
an Machine Learning Learning
strategies in ANNs
165 In artificial Neural Network nodes or neurons weights axons Soma A
interconnected processing
elements are called
166 Each connection link in ANN neurons weights bias activation function B
is associated with
________ which has
information about the input
signal
167 The truth values of traditional Either 0 or 1, Between 0 & 1, either 0 Between 0 & 1, between 0 & 1 Either 0 or 1, either 0 or 1 A
set theory is ____________ between 0 & 1 or 1
and that of fuzzy set is
__________
168 Fuzzy set theory defines set Standard Normal Possibility Poisson C
membership as a -------------
distribution
169 Each element of X is mapped membership value degree of membership membership function both A and B D
to a value between 0 and 1. It
is called _____.
170 Fuzzy Set theory defines AND OR NOT All of the above D
fuzzy operators. Choose the
fuzzy operators from the
following.
171 What are the following Fuzzification → Fuzzification → Rule evaluation → Fuzzification Rule evaluation → A
sequence of steps taken in Rule evaluation → Defuzzification → Rule → Defuzzification Defuzzification →
designing a fuzzy logic Defuzzification evaluation Fuzzification
machine ?
172 Crisp sets are special cases of Simple Sets fuzzy sets Disjoint Sets Union Sets B
….
173 Fuzzy logic is usually IF-THEN-ELSE IF-THEN rules Both IF-THEN-ELSE rules & IF- IF-ELSE rules B
represented as ___________ rules THEN rules

174 Variables whose values are Character Variables Symbolic Variables Linguistic Variables Numeric Variables C
words or sentences in a
natural or artificial language
are known as….
175 Fuzzy to Crisp Conversion is Defuzzification Fuzzification Lambda Set Crispification A
known as
176 The room temperature is hot. Fuzzy Set Crisp Set Normal Set Binary Set A
Here the hot (use of linguistic
variable is used) can be
represented by _______
177 Fuzzification is the process of Vector Scalar Numeric String B
changing ….value into a
fuzzy value.
178 A __________ point of a Core Support Cross-over α - cut C
fuzzy set A is a point x ∈ X at
which µA(x) = 0.5
179 Genetic Algorithm are a part Evolutionary inspired by Darwin's are adaptive heuristic search All of the above A
of Computing theory about evolution - algorithm based on the
"survival of the fittest" evolutionary ideas of natural
selection and genetics
180 Genetic algorithms are TRUE FALSE Some times Can't Say A
heuristic methods that do not
guarantee an optimal solution
to a problem
181 Genetic algorithms belong to Artificial Optimization area Complete enumeration familty Non computer based human A
the family of methods in the Intelligence area methods solutions

182 Which of the following is not It is a probabilistic It is guaranteed to give If an optimization problem has It is an iterative process B
true for Genetic algorithms? search algorithm global optimum more than one solution, then it suitable for parallel
solutions will return all the solutions programming

183 Which one of the following is N , the population the obtainable accuracy the mutation probability the average fitness score D
not necessarily be considered size.
as GA parameters?
184 Which of the following 0-1 Knapsack Travelling salesman Optimal binary search tree Job shop scheduling problem B
optimization problem(s) can problem problem construction problem
be better solved with Order
GA?
185 Which of the following are i & ii i ,ii & iii ii,iii & iv All of the above D
found in genetic algorithm i)
Evolution ii)Selection iii)
Reproduction iv) Mutation
186 Which GA operation is Initial population selection of sub- Reproduction to produce next convergence testing C
computationally expensive creation population for mating generation

187 The pupose of fitness To check whether To decide the To select the best individuals To identify the individual C
evaluation operation all individual termination point with worst cost function
constraints satisfies
the constraints given
in the problem
188 In weighted sum approach, considered converted into a single solved considering one at a time multiplied to get one B
multiple objectives are separately objective using expression for solving
weighting factors
189 Which GA encoding scheme Binary coded GA Real coded GA order GA Tree encoded GA A
gives faster execution?
190 Which of the following GA is deterministic Under any The qualities of solutions offered GAs could be applied to any D
comparison is true? whereas other search circumstances, GAs by GAs for any problems are problem, whereas certain
algorithms are not always outperform other always better than those provided algorithms are applicable to
algorithms. by other search. limited
domains
191 The Fitness Function in method to measure the objective function a substitute to approximate the a least squares A
Genetic Algorithms is how fit a candidate for the optimization survival abilities of individuals in approximation for a
solution is in solving problem being solved nature. polynomial.
the problem.
192 The basic idea behind Genetic of problem solvers of candidate solutions to of candidate solutions in which of problem solvers each of B
Algorithms is to work with a that interact with try and create better each candidate is heuristically which does an independent
population each other through candidates by mixing refined heuristic search
signs genes
193 The basic idea behind Ant of problem solvers of candidate solutions to of candidate solutions in which of problem solvers each of A
Colony Optimization that interact with try and create better each candidate is heuristically which does an independent
algorithms is to work with a each other through candidates by mixing refined. heuristic search.
population signs genes.
194 An ant in Ant Colony a deterministic a stochastic greedy a deterministic perturbation of the a stochastic perturbation of B
Optimization algorithm for greedy constructive constructive method. previous tour. the previous tour
TSP produces a tour by method.
195 What is the relation between The pheromone The pheromone The pheromone deposited on The pheromone deposited on A
the pheromone deposited by deposited on each deposited on each edge each edge is constant. each edge depends upon the
an ant on an edge and the cost edge is directly is inversely proportional length of that edge.
of the tour proportional to the to the cost of the tour
generated by that ant in the cost of the tour
ACO algorithm?
196 A genetic algorithm (GA) for a small population a large population of fit a small diverse population of fit a large diverse population of D
optimization is most likely to of fit and similar and similar individuals. individuals. fit individuals.
succeed given individuals.
197 In the neuron, attached to the Dendrites Axon Synapse Cerebellum A
soma are long irregularly
 shaped filaments called--------
------

198 ADALINE stands for ---------- Adaptive Linear Adaptive Line Neural Adapt Line Neural Element Adaptive Linear Neural A
---------------- Neural Element Network Network Network
Network
199 Self-organizing network also Back Propagation Training free counter Propagation network none B
known as --------------------- network
200 networks were developed by ARP ART ARD none B
carpenter and grossberg
201 ARTMAP is a……..version Unsupervised Semi supervised Supervised Clustering C
of ART
202 Slow learning is used as ------ ART1 ART2 ARTMAP Fuzzy ART B

203 Fuzzy inference also referred approximate reasoning fixed reasoning none A
to as -------------- reasoning
204 Hybrid systems is Genetic Algorithm Genetic Programming Genetic none A
combination of neural
networks, fuzzy logic and -----
---------
205 .------------- deals with neuro –fuzzy neuro-genetic fuzzy –genetic none A
uncertainty problems with its
own merits and demerits

206 .----------------is a neuro-fuzzy Fuzzy ARTMAP Fuzzy art ARTMAP none A

hybrid in which the host is a
recurrent network with a kind
of competitive learning.
207 .----------------- applicable on Fuzzy-genetic neuro – fuzzy fuzzy-logic fuzzy-backpropagation A
fuzzy optimization problems
208 Recurrent network unsupervised hebbian learning reinforced learning supervised learning B
architectures adopting … learning
209 GA-NN also known as -------- GA NNGA GANN none C
---
210 …...maps fuzzy sets and can Fuzzy ART FAN FAM D
encode fuzzy rules
211 Neural network can learn testing training learning none B
various tasks from …
212 Population size, Mutation rate central parameters connection parameters null parameters control parameters D
and cross over rate are
together referred to as
213 The number of elements in a modality placiticity Cardinality elasticity C
set is called its -------------
214 Raising a fuzzy set to its concentration intersection conjunction disjunction A
second power is called
215 BAM stands for ---------------- Biconventional Associative Memory Bidirectional Associative none C
Associative Memory Memory
216 Neural network applications -- Pattern Recognition Optimization Problem Forecasting All of the above D
---------------