M
. Sc.
(Semester I
V
) Examination 200
6

200
7
BIOINFORMATICS
MBIF

4
01
B
:
SOFT COMPUTING TECHNIQUES
Time:
Three
hours Maximum Marks: 70
NOTE:
Answer
Five
Questions
in all
,
including
at least on
e q
uestion
each
from
the
three
Sections
A, B & C
.
Figures on the right

hand side margin indicate Max Marks for each Question
.
SECTION A
1.
(a)
List the typical steps of a Genetic Algorithm to solve a problem.
What
are the differ
ent
(
8
)
kind
of
Genetic operators
often
used?
(b)
Would it be proper to say that Genetic Algorithm is a Population based
stochastic
(
6
)
method?
Explain. Distinguish between Binary Coded and Real Coded G
.
A.
2.
(
a
)
Expla
in the working of following Chromosome Selection techniques:
(
8
)
Roulette
–
wheel Selection
Tournament Selection
(
b
)
Explain with an example, One

point, Two

point and Uniform crossovers. What is (
6
)
the motivation for using
Crossover operator in a Genetic Algorithm?
3
. (a)
What is Building Block hypothesis? What are the typical problems where Genetic
(8)
Algorithms can be applied? List the important advantages and disadvantages
of
using Genetic Algorithms.
(b)
What is Mutation? What is the purpose of using Mutation operator? How does it
(6)
affect convergence rate of the Genetic Algorithm?
SECTION B
4. (a)
Give a simple model of an Artificia
l Neuron. Explain the different entities
in
the (
8
)
model.
List the important parameters of an Artificial Neural Network.
(b)
What are Backpropagation Neural Networks? Show with appropriate diagram how (6)
Per
ceptron model can be used to solve XOR problem.
5. (a)
Briefly explain the following Learning methods in an Artificial Neural Network: (
8
)
Supervised Learning
Hebbian Learning
Reinforced Lear
ning
(b) Describe the Hopfield Neural Network
Model. How does Learning occur in a Hopfield
(
6
)
Network? What is discrete time Hopfield network?
6. (a)
Briefly explain the following:
(
8
)
Radial Basis Function
Cognitron and Neocognitron Models
(b)
What is Stability

plast
icity dilemma? List the important application domains of
(6)
Artificial Neural Networks.
SECTION C
7.
(a)
Distinguish between Fuzzy and Crisp sets. Explain, with an example, the
Membership (5)
f
unction of Fuzzy S
ets.
(b) Consider the set of people in the following age groups:
(9)
0
–
10
10
–
20
20
–
30
30
–
40
40
–
50
50
–
60
60
–
70
70 and above.
The fuzzy sets young, middle aged and old are represented by following membership
function graphs:
Perform the following operations on fuzzy sets A and B, where A = Fuzzy set of young
people and B = Fuzzy set of middle aged people:
(i)
Compute th
e intersection of Fuzzy sets A and
B.
(ii)
Compute the difference of Fuzzy sets
A and B.
(iii)
Compute the disjunctive sum of Fuzzy sets A and B.
8. (a) What are Fuzzy prepositions? Give an example. What connectives are supported by (8)
Fuzzy logic? Given following Fuzzy prepositions P and Q:
P: Mary is efficient, T(p) =
0.8
Q: Ram is efficient, T(Q) = 0.65
What will be the connected preposition P V Q and P => Q? Also compute T(P V Q)
and T(P => Q).
(b) What is Defuzzification? Explain the following methods of Defuzzification:
(6)
C
entroid method
Centre of Sums (COS) method
Mean of Maxima(MOM) method
9. Write short notes on
any two
of the following:
(7 X 2 = 14)
(i) GA based Backpropagation Network
(ii) Fuzzy Associative Memories
(iii)
Fuzzy Logic controlled GA
Comments 0
Log in to post a comment