MCQ Questions:
1. The following is the main distinction between supervised and unsupervised learning:
a. Neuron count b. Labeled data use
c. Training speed d. Type of network
2. The primary applications of fuzzy logic are:
a. A. Binary classification; b. Exact reasoning;
b. C. Imprecise and uncertain data. d. Quick calculation
3. In fuzzy logic, the complement of a membership value μ(x) is:
A. 0 B. 1 C. 1−μ(x) D. μ(x)2
4. Which of the following is minimized in backpropagation?
A. The quantity of layers B. The role of activation
C. The function of error D. Rate of learning
5. In fuzzy logic, which of these is not a type of membership function?
A. Triangle B. Rectilinear C. Gaussian D. Trapezoidal
6. The primary applications of fuzzy logic are:
A. Binary classification; B. Exact reasoning;
C. Imprecise and uncertain data. D. Quick calculation
7. In crossoveroperation, the offspring produced:
A) Are exactly the same as one of the parents
B) Are entirely random and independent of the parents
C) Are created by combining genes from both parents
D) Are always identical to the parent with the highest fitness
8. In the context of a genetic algorithm, thefitness functionis used to:
A) Assign random values to the individuals in the population
B) Evaluate the suitability of individuals for reproduction
C) Generate offspring
D) Perform the mutation operation
9. In Rough Set Theory, thelower approximationof a set represents:
A) The set of objects that definitely belong to the target set
B) The set of objects that might belong to the target set
C) The set of objects that do not belong to the target set
D) The set of objects that can be ambiguously classified
1
10. Which of the following operators is used for intersection in fuzzy logic?
A. The highest B. Minimum C. Product D. Sum
11. Of the following, which is NOT a feature of soft computing?
A. Sturdiness B. Accuracy
C. Flexibility D. Resistance to faults
12. For approximate reasoning, which logic is most appropriate?
A. Classical logic B. Boolean logic C. The fuzzy logic D. Logic by
predicate
13. The output of a perceptron is:
A. Always continuous B. A sigmoid function
C. Binary (0 or 1) D. Unbounded
14. The activation function in an artificial neural network adds:
A. The Linearity B. Sound C. Non-linearity D. Postponement
15. Which of the following operators is used for intersection in fuzzy logic?
A. The highest B. Minimum C. Product D. Sum
16. For approximate reasoning, which logic is most appropriate?
A. Classical logic B. Boolean logic C. The fuzzy logic D. Logic
by predicate
17. Which of the following is not a soft computing technique?
A. Neural Networks B. Fuzzy Logic
C. Genetic Algorithms D. Binary Search
18. Which of the following is used to train neural networks?
A. Clustering B. Backpropagation C. Genetic crossover D. Filtering
19. What happens if the mutation rate in a genetic algorithm is too high?
A) The algorithm will converge quickly
B) It will maintain genetic diversity but slow down convergence
C) The population will become too uniform and premature convergence will occur
D) The population will evolve randomly without any improvement in the solution
20. Which of the following crossover methods uses the concept of exchanging sub-parts
of the solution between two parents?
A) One-point crossover
B) Uniform crossover
C) Multi-point crossover
D) Two-point crossover
2
21. The following factors influence genetic algorithms:
A. artificial intelligence; B. Mendelian heredity;
C. genetics and natural selection; D. quantum physics.
22. In a fuzzy-rough set hybrid, the fuzzy set helps to:
A) Convert crisp attributes to fuzzy values for easier decision-making
B) Reduce the dimensionality of the data
C) Increase the computational complexity
D) Provide exact decision rules without any ambiguity
23. Which of the following operators is used for intersection in fuzzy logic?
A. The highest B. Minimum C. Product D. Sum
24. In fuzzy logic, the complement of a membership value μ(x) is:
A. 0 B. 1 C. 1−μ(x) D. μ(x)2
Short/Long Questions:
1. What distinguishes fuzzy logic from classical logic (crisp set)? Explain the basic
fuzzy set operations: AND, OR, and NOT
2. Write any two fuzzy IF-THEN rules related to temperature control.
3. What is fuzzy logic? Explain its importance in decision-making systems.
4. Describe the process of defuzzification and name two common methods used.
5. What is mutation in genetic algorithms? Why is it important for the algorithm's
performance?
6. Explain how fitness values affect the behavior of Roulette Wheel Selection.
7. Explain the concept of crossover in genetic algorithms. How does it help in
generating new solutions?
8. Define the Discernibility Function. How is it related to the Discernibility Matrix?
9. Explain the Roulette Wheel Selection method used in Genetic Algorithms.
Why is it also known as fitness proportionate selection?.
10. What is the disadvantage if one individual has a fitness much higher than the others?
Given the parent chromosomes:
Parent 1 = 110010, Parent 2 = 101101
3
a) Perform a single-point crossover after the 3rd bit.
b) Write down the resulting offspring chromosomes.
11. Explain the concept of mutation in genetic algorithms. How does it help in
generating new solutions?
12. What is the disadvantage if one individual has a fitness much higher than the others?
Given the parent chromosomes:
Parent 1 = 110010, Parent 2 = 101101
a) Perform a single-point crossover after the 3rd bit.
b) Write down the resulting offspring chromosomes.
13. Describe the process of defuzzification and name two common methods used.
14. Imagine you're in a smart home with a thermostat that uses a fuzzy logic controllerto
regulate the room temperature. The thermostat has fuzzy rules such as:
If the room is very cold, then the heating should be very high.
If the room is slightly warm, then the heating should be moderate.
If the room is hot, then the heating should be low.
Currently, the room temperature is measured as "slightly warm". According to the
fuzzy rules and membership functions, the thermostat gives a fuzzy output that
suggests the heating should be a mix of moderateandlow heating levels, with some
degree of uncertainty in the exact output.Now, the fuzzy controller needs to defuzzify
this fuzzy output into a crisp control value that can be used by the heating system to
adjust the temperature.
How would the fuzzy system defuzzify the output in this scenario, and what
methodwould be used to get a crisp value for the heating level?
15. Discuss how population size affects the performance of a genetic algorithm.
What are the consequences of using:
a) A very small population
b) A very large population
16. Explain two methods to generate the initial population in a genetic algorithm.
17. For the fitness function f(x)=x2, where x is a 5-bit binary number (i.e., 0≤x≤31):
a) Convert the binary chromosome 10110 to decimal.
b) Calculate its fitness value using the given function.
18. Discuss how lower and upper approximations are used in rule generation within rough
set theory.What is the role of reductsinrule generation in rough set theory?
4
19. What is binary encoding in the context of Genetic Algorithms?Explain how binary
encoding can be used to represent a solution for the Knapsack Problem in a Genetic
Algorithm.
20. Why is fuzzification required in fuzzy systems, and what does it entail?
21. Imagine you have a smart fan at home that adjusts its speed based on the room
temperature. The fan uses fuzzy logic to decide how fast it should run based on the
current temperature. The temperature is measured in degrees Celsius.
The fuzzy system uses the following temperature ranges:
1. Low temperature: 15°C to 20°C
2. Medium temperature: 20°C to 30°C
3. High temperature: 30°C to 40°C
When the temperature is 24°C, the fuzzy controller needs to fuzzify this input value
into the fuzzy sets of Low, Medium, and High temperature.
How would the system fuzzify the room temperature of 24°C into the fuzzy sets of
Low, Medium, and High temperatures? What would the membership values for each
fuzzy set be?
22. Define population diversity in the context of genetic algorithms.
Why is it important, and what are two ways to maintain diversity during evolution?
23. What is elitism in genetic algorithms?
24. A company wants to solve a binary optimization problem using a genetic algorithm.
Design a chromosome of length 6 bits to represent a potential solution.
a) Give an example chromosome.
b) What does each bit represent, and why is binary encoding used in GAs?
25. In the context of genetic algorithms, what is binary encoding? Provide a binary
encoding example for a decimal number. Describe the genetic algorithm's binary
decoding procedure.
26. You are solving a problem where the variable xxx ranges from 0 to 63.
a) What is the minimum length of the binary chromosome required?
b) Encode the value x=37 in binary
27. Convert the following decimal numbers to 6-bit binary representation:
a) 12
b) 25
c) 5
Explain each step of the conversion.
