Seat No.: Enrolment No. GUJARAT TECHNOLOGICAL UNIVERSITY BE —SEMESTER- VII EXAMINATION-SUMMER 2023 Subject Code: 3170724 Date: 21/06/2023 Subject Name: Machine Learning Time: 10:30 AM TO 01:00 PM Total Marks: 70 Instructions: 1, Attempt all questions, 2. Make suitable assumptions wherever necessary. 3. Figures to the right indicate full marks. 4, Simple and non-programmable scientific calculators are allowed. MARKS Q.1 (a) Compare Different types of Machine Learning 03 (b) What is Machine Learning? List out its applications and 04 possible ethical issues of machine learning applications? (©) What do you mean by a well-posed learning problem? Explain 07 important features that are required to well-define a learning problem. Q.2 (a) Explain the concept of penalty and reward in reinforcement 03 learning. (b) What is outlier? How can we take care of outliers? 04 (c) Consider the following confusion matrix of the win/loss. 07 prediction of cricket match. Calculate model accuracy and error rate, sensitivity, precision, F-measure and kappa value for the same. ‘Actual Win ‘Actual Loss Predicted Win 85, 4 Predicted Loss [2 9 OR (©) While predicting malignancy of tumor of set of patients using 07 a classification model, following are the data recorded: (a) Correct predictions — 15 malignant, 75 benign (b) Incorrect predictions — 3 malignant, 7 benign Calculate the model accuracy, error rate, Kappa value, sensitivity, precision, and F- measure of the model. Q3° (a) What are the basic data types in machine learning? Give an 03 ‘example of each one of them. (b) Explain the process of K-fold-cross-validation method 04 (©) Explain with an example, main underlying concept of feature 07 extraction. What are the most popular algorithms of feature extraction, briefly explain any one. OR Q3° (a) What are the different techniques for data pre-processing? 03 Explain in brief (b) What is sampling? Explain Bootstrap sampling. 04 (c) What is feature selection? Why it is needed? What are the 07 different approaches of feature selection, briefly explain any one.Q4 Qs Qs (a) (b) © (a) (b) ©) (a) (b) @ (a) (b) © What is joint probability? What is its formula? What is likelihood probability? Give an example Explain the Apriori algorithm for association rule learning with an example, oR Define probability of union of two events with equation, What is concept learning? Explain with example What is supervised leaning? Draw and explain classification steps in detail Explain dependent variable and an independent variable in a linear equation with example Explain Decision tree algorithm Explain in detail, the backpropagation algorithm, What are the limitations of this algorithm? oR Write the strength and weakness of decision tree method. Explain K-mean clustering algorithm What are the different types of activation functions popularly used? Explain each of them, seeeeeeeeseee 03 07 03 07 03 07 03 07Seat No.: Enrolment No. GUJARAT TECHNOLOGICAL UNIVERSITY BE - SEMESTER-VII (NEW) EXAMINATION — WINTER 2022 Subject Code:3170724 Date: 16-01-2023 Subject Name:Machine Learning Time:10:30 AM TO 01:00 PM. Total Marks:70 Instructions: 1. Attempt all questions. 2. Make suitable assum \s wherever necessary. 3. Figures to the right indicate full marks. 4, Simple and non-programmable scientific calculators are allowed. MARKS Q (a) Define Machine learning and list out few applications in Engineering? 03 (b) Distinguish lazy vs eager learner with an example. 04 (©) Elaborate the cross validation in training a model, 07 Q2 (a) Whatis conditional probability? Define its importance. 03 (b) What is categorical data? Explain its types with examples. 04 (©) What is principal component analysis? How does it work? Explain. 07 OR (©) What is model accuracy in reference to classification? Also Explain the 07 performance parameters Precision, Recall and F-measure with its formula and example. Q3 (a) Define the following terms. 03 @) Variance (ii) Covariance Gi) Joint Probability (b) What is the purpose of Singular value decomposition? How does it 04 achieve (ce) Explain KNN algorithm with suitable example. Oy OR Q3 (a) Explain posterior probability with its formula. 03 (b) Show various distance-based similarity measure with its example. 04 (©) Define feature and explain the process of transforming numeric features 07 to categorical features with suitable example. QA (a) What is Bernoulli distribution? Explain briefly with its formula. 03 (b) Explain the concept of Bayesian belief network. 04 (©) Explain decision tree approach with suitable example. 07 OR Q.4 (a) Mention few applications areas of unsupervised learning in Engineering. 03 (b) Define Entropy. Show its importance with suitable example. 04 (©) How does the apriori principle help in reducing the calculation overhead—(07 for a market basket analysis? Explain with an example. Q.5 (a) Briefly explain Perceptron and Mention its limitation. 03 (b) Define linear regression. Also explain Sum of squares with its formula. 04 (©) Explain k-means clustering technique. 07 OR QS (a) What are the strengths and weaknesses of SVM? 03(b) Briefly explain K-Medoids. (©) Show the Step, ReLU and sigmoid activation functions with its equations and sketch. 04 07Seat No.: GUJARAT TECHNOLOGICAL UNIVERSITY BE - SEMESTER-VII (NEW) EXAMINATION ~ WINTER 2021 Subject Code:3170724 Subject Name:Machine Learning Time Instructions: Qu Q2 Q3 Q3 Qa Qa 1 2 3. 4 (a) (b) «© (a) (b) © (©) (a) (b) © (a) (b) cC) (a) (b) © (a) (b) © Attempt all questions, 0:30 AM TO 01:00 PM Make suitable assumptions wherever necessary. Figures to the right indicate full marks. simple and non-programmable scientific calculators are allowed. Give the difference between supervised leaning and unsupervised learning Explain the concept of penalty and reward in reinforcement learning. What do you mean by a well-posed learning problem? Explain important features that are required to well-define a learning problem. How can we take care of outliers in data? Consider the following confusion matrix of the win/loss prediction of cricket match. Calculate model accuracy and error rate for the same. ‘Actual Win ‘Actual Loss Predicted Win 82 7 Predicted Loss 3 8 Explain SVD as a feature extraction technique with suitable example. Explain K-fold cross va In a communication system each data packet consists of 1000 bits. Due to the noise, each bit may be received in error with probability 0.1. It is assumed bit errors occur independently. Find the probability that there are OR idation method with suitable example. 13% of electronic units manufactured by a company are defective. Find the probability that in a sample of 200 units, less than 2 bulbs are defective. Explain how Naive Bayes classifier Discuss appropriate problems for decision tree learning in detail OR more than 120 errors in a certain data packet. What is likelihood probability? Give an example. Discus example. Describe the concept of single link and complete link in the context of hierarchical clustering. Explain how the Market Basket Analysis uses the concepts of association analysis. Explain dependent variable and an independent variable in a linear equation with example. Describe the main difference in the approach of k-means and k-medoids algorithms with a neat diagram, Explain the Apriori algorithm for association rule learning with an example the error rate and validation error in the KNN algorithm, Explain sum of squares due to error in multiple linear regression with OR used for Spam Filtering. Date:17/12/2021 Total Marks: 70 MARKS 03 07 07 03 07 03 07 03 04 07 03 04 07Qs (a) (b) © (a) (b) Co) ions of a negative slop. Explain Rosenblatt’s perceptron model. Describe, in details, the process of adjusting the interconnection weights in a multi-layer neural network. linear regression? oR ‘What are the factors determining the effectiveness of SVM? Draw a flow chart which represents backpropagation algorithm, Explain, with example, the challenge in assigning synaptic weights for the interconnection between neurons? How can this challenge be addressed? 03 07 03 07GUJARAT TECHNOLOGICAL UNIVERSITY BE - SEMESTER-VII (NEW) EXAMINATION - SUMMER 2024 Subject Code:3170724 Date:15-05-2024 Subject Name:Machine Learning Time:02:30 PM TO 05:00 PM. Total Marks:70 Instructions: 1, Attempt all questions, 2. Make suitable assumptions wherever necessary. 3. Figures to the right indicate full marks. 4, Simple and non-programmable scientific calculators are allowed. Q.1 (a) What is reinforcement learning? State key features of reinforcement learning. (b) Compare the different types of machine learning, (©) What is IQR? How is it measured? Explain with proper example. Q.2 (a) How does discretization help us to deal with outliers’ (b) Explain qualitative and quantitative data. (©) What are overfitting and underfitting in context of machine learning models? What is the major cause of overfitting and underfitting? OR (c)_ How to improve the performance of a learning model? Explain in details. Q3 (a) What is feature engineering? (b) Define sampling with replacement. Give proper example. (©) Explain naive bayes classifier with an example. OR Q.3. (a) When can a feature be termed as irrelevant? How can it be measured? (b) What is sampling without replacement? Give proper example. (©) Explain Bayesian belief network. ferentiate between the two. Q4 (a) Explain the classification learning steps. (b) What are the factors determining the effectiveness of SVM? (©) Discuss the decision tree algorithm in detail OR Q.4 (a) Why the KNN algorithm is called a lazy learner? (b) Describe the strengths and weaknesses of decision tree. (©) Explain the difference between linear and logistics regression with example. Q5_ (a) Discuss the strengths and weaknesses of the k-means algorithm. (b) Explain hierarchical clustering. (©) Explain the different types of activation fun OR Q5_ (a) Differentiate between supervised learning versus unsupervised learning. (b) Explain density-based methods. (©) Explain the backpropagation algorithm. What are the limitations of this algorithm? ns used in neural network. Marks 03 07 03 07 07 03 07 03 07 03 07 03 07 03 07 03 07Seat No.: Enrolment No. GUJARAT TECHNOLOGICAL UNIVERSITY BE - SEMESTER-VII (NEW) EXAMINATION ~ SUMMER 2022 Subject Code:3170724 Date:14/06/2022 Subject Name:Machine Learning Time:02:30 PM TO 05:00 PM Total Marks: 70 Instructions: 1. Attempt all questions. wherever necessary. te full marks, 4. Simple and non-programmable scientific calculators are allowed. Q.1_ (a) _ Define Machine learning? Briefly explain the types of learning, 03 (b) Explain the concept of penalty and reward in reinforcement. Learning. 04 (©) What do you mean by a well-posed learning problem? Explain important 07 features that are required to well-define a leaming problem, Q2 (a) How can we take care of outliers in data? 03 (b) Explain Key elements of Machine Learning. Explain various function 04 approximation methods. (©) Draw and explain the flow diagram of machine learning procedure. 07 OR (©) List and explain the types of machine learning in brief. 7 Q.3. (a) What is likelihood probability? Give an example. 03 (b) What is data sampling? Explain data sampling methods? 04 (©) What are the Techniques Provided in Data Preprocessing? Explain in 07 brief. OR Q3° (a) What is difference between Machine Learning and Deep Learning. 03 (b) Differentiate PCA and LDA. 04 (©) Explain the process of Supervised Learning Model 07 Q4 (a) _ Define issues in machine Learning. 03 (b) Write a note on KNN. 04 (c) List the methods for Model evaluation. Explain each. How we can 07 improve the performance of model. OR Q4 (a) Explain the training of Predictive Model. 03 (b) List Classification algorithms. Explain Decision Tree as 04 classification method. (©) What is Clustering? Explain K-mean clustering algorithm. 7 (a) Explain the need of feature engineering in ML. 03 (b) Explain Binomial Distribution with an example. 04 (©) Explain Bayes’ theorem in details. 07 OR Q5 (a) Define: 03 a. Supervised Learning b. Classification c. Regression (b) Write a short note on feed forward neural network. (©) Explain Monte Carlo Approximation. SfSeat No.: Enrolment No. GUJARAT TECHNOLOGICAL UNIVERSITY BE - SEMESTER-VII (NEW) EXAMINATION — WINTER 2023 Subject Code:3170724 Date:01-12-2023 Subject Name: Machine Learning Time: 10:30 AM TO 01:00 PM Total Marks:70 2. Make suitable assumptions wherever necessary. 3. Figures to the right indicate full marks. 4. Simple and non-programmable scientific calculators are allowed. MARKS Q.1 (a) Differentiate Human Learning & Machine Learning 03 (b) What is Feature Subset Selection? Explain it in detail 04 (©) Explain Supervised Learning in detail. 07 Q2 (a) What is K-fold cross validation? 03 (b) What is Clustering in Unsupervised Learning? Explain it in detail 04 (©) Explain Reinforcement Leaning in detail. 07 OR (©) Explain dimensionality reduction with its advantages and 07 disadvantages. Q3 (a) _List out different issues in machine Learning. 03 (b) Explain Standard Deviation and Variance with an example in machine 04 learning. (©) What are the techniques provided in Data Pre-processing? Explain it 07 in brief. OR Q3 (a) What is the difference between Qualitative Data and Quantitative 03 Data? (b) Explain Under-fitting and Over-fitting with an example. 4 (©) Explain Probability Distribution Function and Cumulative 07 Distribution Function with an Example. Q.4 (a) What is Multiple Linear regression? 03 (b) What is a bad decision boundary in SVM? Explain it in brief. 04 (©) Explain Decision Tree in details. 07 OR Q4 (a) What is Monte Carlo Approximation? 03 (b) Explain Apriori algorithm with its Limitation. 04 (©) Explain Sampling Distribution with different types of Sampling. 07 Q5 (a) What is Deep Learning? 03 (b) Explain Hypothesis testing with basic assumptions. 04 (©) Explain k-Nearest Neighbor classification method in detail with its 07 advantages and disadvantages. oR Q5 (a) Explain Forward pass and Backward pass in neural network. 03 (b) List out the characteristics of Artificial Neural Network and explain it 04 in brief. (©) Explain Bayes’ theorem in details with the use of Bayesian classifiers. 07 eeemeneneeeee