CHAPTERS CHP 1 CHP 2 CHP 3 CHP 4 CHP 5 CHP 6
QUESTIONS
QUESTION
NO 1
QUESTION
NO 2
QUESTION
NO 3
QUESTION
NO 4
QUESTION
NO 5
QUESTION
NO 6
M25
Q.1 Answer the following (Any four)
a. Implement AND function using Mc-Culloch-Pitts neuron.
b. Discuss the limitations of deterministic approaches and how probabilistic reasoning addresses these limitations
with example.
c. Discuss different activation function used in Neural Network.
d. Compare between ANN and RNN.
e. Describe the role of utility and expected utility in decision theory with example.
f. Differentiate between Fuzzyness and Probability.
Q.2
a. Explain the component of ANN architecture.
b. Explain the role of representing knowledge in taxonomies and ontologies, and how advanced analytics can be
applied to cognitive systems.
Q.3
a. Explain different types of membership functions used in fuzzy logic with suitable examples. Differentiate
between crisp sets and fuzzy sets in terms of their structure, interpretation, and applicability in real-world
scenarios.
b. Using Mamdani fuzzy model design a fuzzy model controller to determine the wash time of domestic washing
machine. Assume the inputs are dirt & grease on cloths. Use three descriptors for each input variable & five
descriptors for output variables. Derive necessary membership functions & required fuzzy rules for the
application.
Q.4
a. Explain Architecture of CNN in detail. List any two applications of it.
b. What is Bayesian Belief Network? Explain inferencing with example.
Q.5
a. Explain any 4 Metrics for evaluating classifier performance. Discuss any two cross validation methods.
b. How Bagging and Boosting handle bias-variance trade-off differently, and analyze their effectiveness in dealing
with noisy data and overfitting. Explain with algorithmic such as Random Forest and AdaBoost.
Q.6
a. Explain the role of machine learning in multimodal applications. Discuss how different data modalities (text,
audio, image, and video) are integrated to improve the performance and user experience of these systems with
real time applications.
b. Define Defuzzification? Explain any two methods of Defuzzification.
D24
Here are the questions from the paper in a clean text format:
Q.1 (5 marks each)
(a) Explain Bayes theorem.
(b) Consider two fuzzy sets:
 = {0.3/1 + 0.4/2 + 0.5/3 + 0.6/4}
Ê = {0.1/1 + 0.2/2 + 0.2/3 + 0.1/4}
Find the algebraic sum, algebraic product, bounded sum, and bounded difference of the given fuzzy sets.
(c) Explain Mc-Culloch-Pitts neuron with an example.
(d) Explain bootstrap for sampling.
Q.2 (10 marks each)
(a) Explain Ensemble Methods.
(b) Define Cognitive Computing. Draw a neat diagram of components of the cognitive system and explain the
components.
Q.3 (10 marks each)
(a) Explain the components of ANN architecture.
(b) Perform a case-study on video recommendation system (data science based).
Q.4 (10 marks each)
(a) Define Defuzzification. Discuss any two methods of defuzzification?
(b) What is the Bayesian Belief Network? Illustrate with an example.
Q.5 (10 marks each)
(a) Describe Natural Language Processing in Support of a Cognitive System.
(b) Explain in detail the Long Short-Term Memory Network with an example.
Q.6 (10 marks each)
(a) Define Accuracy, precision, and recall. Evaluate performance of classifier1 and classifier 2 on the basis of above
evaluation parameters, given the following confusion matrices:
Classifier 1
F' N'
F 20 10
N 10 60
Classifier 2
F' N'
F 0 15
N 5 80
(b) Write a short note on: Trends in Data Science for audio.
M24
Q.1
(a) From a standard deck of playing cards, a single card is drawn. The probability that the card is king is 4/52, then
calculate posterior probability P(King|Face), which means the drawn face card is a king card.
(b) Describe the Centroid method of defuzzification. Include its formula and discuss its advantages and
disadvantages compared to other defuzzification methods.
(c) List down different applications of Deep Learning. And explain it.
(d) Compare and contrast Bagging and Boosting with their application.
Q.2
(a) How do you explain random forest? Does random forest need pruning, explain in detail?
(b) Describe how cognitive computing can be applied in healthcare. Provide specific examples of tasks or problems
that cognitive computing can help address, and explain the potential benefits.
Q.3
(a) Explain how an autoencoder can be used for dimensionality reduction. Include a brief description of how the
encoder and decoder parts of the network work in this context.
(b) List three industries where data science is extensively used and briefly describe one application in each
industry.
Q.4
(a) Using Mamdani fuzzy model design a fuzzy logic controller to determine the wash time of domestic washing
machine. Assume the inputs are dirt and grease on clothes. Use three descriptors for each input variable and five
descriptor for output variables. Derive necessary membership functions and required fuzzy rules for the
application.
(b) What is a Markov Decision Process (MDP)? List its primary components. Explain in detail.
Q.5
(a) What is cognitive computing, and how does it differ from traditional computing? List the process of building a
cognitive application.
(b) Describe the architecture of a typical CNN. What are the main components, and how do they contribute to the
network's performance in image recognition tasks?
Q.6
(a) Let's consider a binary classification problem where we have built a classifier to predict whether a transaction
is fraudulent (positive class) or legitimate (negative class). After training the classifier and testing it on a dataset,
we obtain the following confusion matrix:
Predicted Legitimate Predicted Fraudulent
Actual Legitimate 850 20
Actual Fraudulent 30 100
Calculate Accuracy, Precision, Recall and F1-score.
(b) Briefly explain Data science for Multimodal applications.
D23
Q.1
(a) From below given probability distribution find P(¬Cavity | Toothache)
(b) Define defuzzification and State the necessity of the defuzzification process.
(c) Implement AND function using Mc-Culloch-Pitts neuron. (take binary data)
(d) What is the significance of ROC curves?
Q.2
(a) State Ensemble methods and describe anyone.
(b) Illustrate usage of taxonomies and ontologies for knowledge representation in cognitive system.
Q.3
(a) Explain the components of CNN architecture.
(b) Perform a case study on book recommendation system (data science based)
Q.4
(a) Describe the Properties of Fuzzy Sets with an example.
(b) Illustrate inferencing in Bayesian Belief Network with an example.
Q.5
(a) List and explain the design principles of Cognitive System.
(b) State and elaborate the applications of deep learning.
Q.6
(a) Calculate Accuracy, Precision, Recall, Sensitivity and Specificity for the following example:
Actual Class \ Predicted Class Buys_Computer=yes Buys_Computer=no
Buys_Computer=yes 6954 46
Buys_Computer=no 412 2588
(b) Write a short note on- Data Science for Multi modal applications.
M23
Q.1
(a) Which are the various metrics for evaluating the classifier performance?
(b) Briefly explain CNN architecture in detail.
Q.2
(a) Explain the XOR function using McCulloch-Pitts model (use binary data representation).
(b) How do you explain random forest? Does random forest need pruning, explain in detail?
Q.3
(a) What is the need of Markov Decision Processes? Explain MDP in detail.
(b) Explain the process of building a cognitive application.
Q.4
(a) Using Mamdani fuzzy model design a fuzzy logic controller to determine the wash time of a domestic washing
machine. Assume the inputs are dirt and grease on clothes. Use three descriptors for each input variable and five
descriptors for output variables. Derive necessary membership functions and required fuzzy rules for the
application.
(b) What are the essential components of an Autoencoder? Explain in detail. Which activation function is best for
autoencoder?
Q.5
(a) How does class imbalance affect classification? What are the ways to solve the class imbalance problem?
(b) Explain Data science for Multimodal applications.
Q.6 Write Short Note on Any Four
(a) Holdout method
(b) LSTM
(c) Defuzzification Methods
(d) Design Principles for Cognitive Systems
(e) Bayes Theorem
(f) Application of Data science for text