AKTU MCA Soft Computing (KCA032) Question Paper 2024-25

Note: Attempt all Sections. In case of any missing data; choose suitably.

Section A

Q1. Attempt all questions in brief · 2 × 10 = 20 marks

1. What are the key differences between soft computing and hard computing?
2. Define an artificial neuron and describe its basic structure.
3. What is the role of activation function in a neural network?
4. Explain the concept of fuzzification and its purpose.
5. List different types of learning in neural networks.
6. What are the key operations in a genetic algorithm?
7. Define membership function in fuzzy logic.
8. What is crossover in genetic algorithms and why is it important?
9. Explain the term ‘swarm intelligence’.
10. What are hybrid soft computing techniques? Give an example.

Section B

Q2. Attempt any three of the following · 10 × 3 = 30 marks

1. Critically evaluate how the choice of activation function affects the learning capability, convergence, and performance of a neural network in real-world applications such as image recognition or language processing. Justify your reasoning with suitable examples.
2. Explain the architecture and training process of multilayer perceptron.
3. Various Fuzzy set operations extend classical set theory to handle uncertainty and vagueness. Critically analyze how these operations enable better modeling of real-world problems compared to crisp logic. Choose a real-life scenario (e.g., air conditioner control, risk assessment, or medical diagnosis) and demonstrate how fuzzy set operations enhance decision-making in that context.
4. Explain the structure and working of a Kohonen Self-Organizing Map.
5. Explain the concepts of Genetic Algorithm. Analyze and evaluate how each step of a GA—selection, crossover, and mutation—contributes to finding optimal or near-optimal solutions. Also, discuss potential challenges in applying GAs effectively.

Section C

Q3

Attempt any one part of the following · 10 × 1 = 10 marks

1. Explain supervised, unsupervised, and reinforcement learning with suitable examples.
2. Critically analyze how soft computing differs from hard computing in dealing with real-world complexity. Choose a practical application domain (such as medical diagnosis, autonomous vehicles, or financial prediction) and argue why soft computing is better suited than traditional hard computing approaches.

Q4

Attempt any one part of the following · 10 × 1 = 10 marks

1. Compare and contrast Perceptron and Backpropagation networks. Also, Design a perceptron that implements the logical AND function. Provide the weight updates for each training step using a learning rate of 1.
2. Explain the architecture and features of a Hopfield Network.

Q5

Attempt any one part of the following · 10 × 1 = 10 marks

1. Explain fuzzy inference system with a diagram. Describe its components. Also, Given fuzzy set A = {(1, 0.3), (2, 0.5), (3, 0.7), (4, 1)}, compute the α-cut for α = 0.6 and the complement of A.
2. Differentiate between Mamdani and Sugeno inference systems.

Q6

Attempt any one part of the following · 10 × 1 = 10 marks

1. Explain the process of encoding, crossover, and mutation in GA with examples. Also, Apply single-point crossover and mutation on: Parent 1: 11010101 Parent 2: 10101011 Crossover after 4th bit; flip 3rd bit of offspring.
2. Compare and contrast Genetic Algorithms (GAs), Genetic Programming (GP), and Evolutionary Strategies (ES) in the context of solving real- world optimization problems. Choose one real-world problem (e.g., route optimization, financial forecasting, robotic control, etc.) and justify which of the three would be the most appropriate approach and why.

Q7

Attempt any one part of the following · 10 × 1 = 10 marks

1. Describe the neuro-fuzzy hybrid system with a suitable application.
2. Swarm Intelligence algorithms such as Particle Swarm Optimization (PSO) and Ant Colony Optimization (ACO) are inspired by collective behavior in nature. Critically evaluate how these algorithms mimic real- world swarm behavior to solve optimization problems. Choose a specific application (e.g., network routing, task scheduling, or vehicle routing) and analyze the advantages and limitations of using swarm intelligence for that problem.

Other years

More AKTU Soft Computing papers on LearnAIFast.

This paper is reproduced for study purposes only. Copyright belongs to AKTU. Source: the question paper bank of ABES Institute of Technology, an AKTU-affiliated college. If you own the copyright and want it removed, email learnwithskillnest@gmail.com and we will take it down.