l o

DAYANANDA SAGAR UNIVERSITY

USN No:
z017
2019/June
Examinations
May Course Code: 16CS344

V Semester B. Tech D a t e : 2 0 - 0 5 - 2 0 1 9

Course Title: Soft Computing Max Marks: 60

Duration: 03 Hours
Time: 10:00 AM to 01:00 PM
Section
one from each
choosing
Note: .Answer 5 full questions
Marks
2. Each Section carrles 12
wherever
necessary
sketches
3. Draw neat assumed
be sultably
4. Missing Data may
SECTION-1
(08 Marks)
example.
with an
Explain with bias
How to design a learning system? below figure (02 Marks)
1.a. shown in
network

Calculate the net input for

1.b.
included in the network. 1

0.3
0.43
0.4( )

o.7 (02 Marks)

computing.
List any two
applications ofsoft
1.c.
OR
(06 Marks)
computing
and hard computing
soft
Compare
What is soft
computing?
to specific ordering
general (06 Marks)
2.a. Give an example for
learning?
What is concept
2.b.
of hypothesis.
SECTION-2

(06 Marks)
reinforcement learning.
unsupervised
and
contrast supervised, (06 Marks)
Compare and
3.a. network.
learning rule for neural
Hebb
State and explain
3.b.

OR (08 Marks)
Culloch-Pitts n e u r o n .
function using MC (P.T.0)
4.a. Implement XOR

105
 (04 Marks)
example.
4.b. Define linear seperability with
SECTION-3

algorithm and
with
network

5.a. Explain the working of Back

Propagation (08 Marks)
Adaline
architecture diagram. interconnections
between

5.b. the characteristics of

weighted (04 Marks)
Ainalyze
and Madaline network.
OR
Tree Learning. (08 Marks)
in Decision
different issues
6.a. in detail about
Explain (04 Marks)
.0. Write a short note on Associativememory

SECTION-4

conventional algorithm? Give

differ from
7.a. How does genetic algorithm (06 Marks)
conventional algorithm.
advantages of GA over
basic structure
of genetic algorithm. (06 Marks)
7.b. With a neat diagram, explain the
OR

mutation functions in genetic

algorithm. (05 Marks)
8.a. Explain the different types of
what are the selection
8.b. Under the reproduction phase of genetic algorithm
detail and which one is most effective
techniques available explain in
among them. (07 Marks)

SECTION 5

Consider the following sentences:

9.a.
Marcus was a man i) Marcus was a Pompeian

ii) All Pompeian's were romans iv) Everyone is loyal to someone

Translate above sentence to predicate logic. (04 Marks)
9.b. Identify the method used for effective implementation of learn one rule.
Explain in detail. (08 Marks)

OR

10.a. Consider the problem of finding the shortest route through several cities,
such that each city is visited only once and in the end return to the starting
city. Which soft computing technique is used to solve this problem and
how it is solved? (08 Marks)

10.b. Write a short note on Neuro fuzzy systems. (04 Marks)

108
 113 tUni
L 1BRARY1|
DAYANANDA SAGAR UNIVERSITY
Banglore
a 5

USN No:
VI Semester B. Tech Examinations - May 2018

Course Title: Soft Computing Course Code: 15CS344

Duratlon: 03 Hours Date: 30-05-2018
Time: 10:00 AM to 01:00 PM Max Marks: 70
Note: 1. Answer 5 full questlons choosing one from each Section
2. Each Section carries 14 Marks
3. Draw neat sketches wherever necessary
4. Missing Data may be sultably assumed

SECTION-1

1.a. What is soft computing? Explain the importance of softcomputing (05 Marks)
1.b. What is Fuzzy computing? How do you capture uncertainty using fuzzy?
Explain with example. (05 Marks)
1.c. Explain with an example, the Graphical Interpretation of fuzzy set for
PRIME numbers. (04 Marks)

OR

2.a. What are the goals of soft computing? List down its applications. (04 Marks)
2.b. Explain the following Fuzzy operations with example: (06 Marks)
) Inclusion (i) Equality ii) Complement
2.c. What is Genetic Algorithm? Explain the basic concepts of Genetic Algorithm. (04 Marks)

SECTION-2

3.a. Represent the ANN mathematical models along with neat diagram for: (06 Marks)
Piece wise linear function (i) Sigmoidal function

3.b. Design the architecture of Multi-Layer Feed Forward Network with neat
diagram and explain the concept behind it. (05 Marks)
3.c. Explain the concept of Hebbian Learning in ANN. (03 Marks)

OR

Page 1 of 3

I13
 114
4.a. How do you train the perceptron using learning algorithm? (04 Marks)
the following Training
4.b. now do you train Back Propagation Network for
sets shown in the Table below? (10 Marks)
| SI. No Inputs Output
1 12
0.7 0.1
0.4
2 0.3 -0.5 0.05
0.6 0.1 0.3
3
4 0.2 0.4 0.25
5 0.1 -0.2 0.12

SECTION-3

set theory. (04 Marks)

5.a. Distinguish fuzzy set theory and classical
5.b. Given: Set for Temperatures be X {30, 40, 50,
= 60, 70, 80, 90, 100}
Set of Rotations/Min be Y {10, 20, 30, 40,
= 50, 60} and
H={(70,1) (80, 1) (90, 0.3)}
VH={(90, 0.9) (100, 1)}
QS ={(10,1) (20, 0.8)}
S={(30, 0.8) (40, 1) (50, 0.6)}
QS(Quite Slow) indicate fuzzy
where, H(High), VH( Very High), S(Slow),
sets
rules to deduce Rotation is Quite
Slow. (10 Marks)
Apply the Fuzzy Modus Ponens
Rotation is Slow.
IF Temperature is High THEN the

(i) Temperature is Very High.

OR

considering example fuzzy sets: (04 Marks)

6.a. Apply the following fuzzy properties by
(i) Intersection
() Union
for a Car Speed at 40 km/h. (04 Marks)
6.b. Represent Fuzzyfication
Min-Max Compositions. (06 Marks)
6.c. Demonstrate the fuzzy Relations using

SECTION-4

Methods used in Gas (Genetic

7.a. Explain with neat diagram Optimization
(06 Marks)
Algorithm's).
(04 Marks)
GA.
7.b. Explain Roulette Wheel Selection in
(04 Marks)
in GAs.
7.c. Explain Classical (Calculus) based Search

OR

Page 2 of3

11
 nanda
Sapa
115
LIBRARY
8.a. Give the outine steps of basic Genetic Algorithm (07 Marks)
8.b. Explain Binary Encoding in GA with example.
gaior S60o
(07 Marks)

SECTION-5

9.a. What is Hybridization? Explain the classification of Hybrid Models (08 Marks)
9.b. Explain Fuzzy Associative Memory with example. (06 Marks)

OR

10.a. Explain Supervised Fuzzy ARTMAP. (04 Marks)

10.b. Explain Fuzy-Back propagation Network architecture. (10 Marks)

Page 3 of 3
115
 RANKA
Mod2 DATE
PAGE

Calcuid6 u net thpu tu Cca sawscn

2 , -03
-aw+aa2 F3Wz
Oo2t 6 Sf O O6 -o

Srput wurhia, Dulud

bias irput a-1

b) aob = 0
S
45 b 2x0.3 O.6 xo.
0
6 93

espuk

ell
el Apela e i d a
35
 RANKA

.w

+ 12WO2 Lw
6 3 s - o 8 0 1 + 6 rD 4z-
O55DO 0.180.o Sa

tos bhaa $1amenl a ala

$3
lt

2
L S3

mpl'de ha
T2uh ala

t w iO=1,1)
-n= {,i
 RANKA
DAITE
PAGE

p w wn
NDNOT

LRIe 2

x1-P
2

Cu, w
,)
in IxD lOx-1=4
 -RANKA
DATE

PAGE

Tpmont No wnth' HP
h hue 3R m c oluy 1

A OxL=
Otltx =

t
=Ox/Ix-=-|

w - L 1
O0)
2
 RANKA
A

22

2
 a ubo Kohucr to Înpllnut
uehUbtp uut d fannalz
Logitas AND
Input
Tauyst)
-d
- )
-1

Smal i t s b =D old y alM c w t

W b =o

04o -
C Cnuw) wcod) +X.y
O+ 0).0)

Cw2:(hu) w2 d) + dlz

w(rd wunN
Li

C)
AWr ty -()
 l d wegl u wGAL Le g in_
euhaub Ktep
n y heur

Cla tuwNs
wLeld).[o1d) | w Cnw o1d)) tniy
1t1E)

CWCnece)= weld) + aL

b (-1)

Ah4=

g60 inpuu
wCnt)E 04G)G)_

b +E)

AW2
A WL
AD
 RANKA-
DATE I
PAGE

(a b )4
Cw)=At (-1) G
wu)14(-)A)
2
b 4 -
2

4 F)-) =
AW H) E1)=
Ab

AwAWn 4b w wb
1
-
02
27
ouptsaVny

w
- t

-n, -F)
 RANEA

esluuuer

2 2 2 = u w uglls _bla