Download as pdf or txt
Download as pdf or txt
You are on page 1of 3

Roll No:_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

Amrita Vishwa Vidyapeetham


Amrita School of Engineering, Coimbatore
B.Tech Degree Examination – April 2019
Sixth Semester
Computer Science and Engineering
15CSE338 Computational Intelligence
Time: Three Hours Maximum: 100 Marks

COs Course Outcome


CO 01 Understand the nature and purpose of different computational intelligent components.
CO 02 Study and design artificial neuron and artificial neural network with learning and
adaptations.
CO 03 Design of fuzzy inference system and its implementation for simple decision making
applications with uncertain data.
CO 04 Analyze working of Evolutionary Algorithms and use them for simple optimization
problems.
CO 05 Understanding hybridization approaches for integrating computational intelligent
components.

Part A (10 x 5 = 50 Marks)


1) The neural network shown in the given figure is to be designed to behave as a logic converter.
The node has a binary transfer function. Draw the pattern hyperplane in the weigh space and
shade the solution region. [5] [CO 02]

2) Write the generalized perceptron learning algorithm. Explain its working principle.
[5] [CO 02]

3) What would you expect to happen if the Perceptron learning algorithm is applied to a non-
linearly separble pattern set? [5] [CO 02]

4) Fuzzy sets V and W are defined on the same universe of five individuals as follows
{ } [5] [CO 03]

{ }

Find (a) ̃
(b) ̃
(c)
(d)
(e) ̃ ̃

R Page 1 of 3
5) For the fuzzy sets X and Y defined over the same universe, prove the following with suitable
example [5] [CO 03]
a. If
b. If

6) If R, S and T are three binary fuzzy relations, then prove that Ro(S U T) = (R o S) U (R o T)
using the definition of composition and union operations on fuzzy relations. [5] [CO 03]

7) ABC is a cake maker, trying to optimize the sugar, flour and butter in his cakes. ABC is trying
to find the optimal combination in the three dimensional space. One axis is the number of
kilograms of flour per batch, the other is the number of kilograms of sugar; and the number
grams of butter; the quality of the cake is a function of these three. How many combinations
he has to test for the given case? Will a brute force strategy help him to find optimal solution?
ABC decides to use GA for solving this problem. Design the chromosome structure. ABC
mimics the mutation by selecting one of the chromosome’s genes randomly, and alters it
randomly by adding or subtracting 1, taking care to stay within the range. Show how mutation
is performed with an example. ABC cuts two chromosomes at two points and rejoins them to
get a fortunate combination. What is this process? [5] [CO 04]

8) For the given initial population of 8-queens problem, apply the genetic algorithm and create
the population after three generations. Use your own mutation and crossover strategy.
Population size is 4 and the candidates in the population are [5] [CO 04]

2 4 7 4 8 5 5 2
3 2 7 5 2 4 1 1
2 4 4 1 5 1 2 4
3 2 5 4 3 2 1 3

9) How do you convert the following formulas as candidates for a population to use in genetic
programming? Perform possible mutations and crossovers with your candidate.
1) IF (NOC = 2) AND (S > 80000) THEN good ELSE bad [5] [CO 04]
2) (2.2 – (x / 11))+7*cos(Y))

10) Write the advantages and classifications of hybrid soft computing techniques. [5] [CO 05]

Part B (5 x 10 = 50 Marks)
11) a) In biological neural network, explain the how the action potentials are generated and
propagated. [7] [CO 01]
b) Draw the diagram for a typical synapse in biological neural network. [3] [CO 01]

12) Draw the mathematical model for a Threshold Logic Neuron. Solve the logic XNOR problem
geometrically using a two layer TLN architecture to find appropriate weights and threshold for
different neurons. [10] [CO 02]

R Page 2 of 3
13) Explain the stages of constructing a Mamdani-type fuzzy controller, for the following fyzzy
system. The fuzzy system has to smoothly slow and stop a train, which is travelling at any
speed and is at any distance from the station. Assuming the train is moving at 3 km/hr and is
at 1.8 m from the station, determing the percentage of braking to be applied. [10] [CO 03]

14) (a) Compare the operations of classical sets and fuzzy sets. [4] [CO 03]
(b) Write the properties of classical and fuzzy relations in both set-theoretical format and
function-theoretical formats. [6] [CO 03]

15) a) Explain the two situation where selection of candidates in done during the evolution of an
evolutionary algorithm. [4] [CO 04]

b) Considering the following candidate solutions in a population under Genetic Algorithm


demonstrate various mutation and crossover schemes available in Genetic algorithms.
[6] [CO 04]
1 2 3 4 5 6 7 8
11 12 13 14 15 16 17 18

*****

R Page 3 of 3

You might also like