Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

SRM VALLIAMMAI 1924103-Machine-Learning

Download as pdf or txt
Download as pdf or txt
You are on page 1of 10

SRM VALLIAMMAI ENGINEERING

COLLEGE
(An Autonomous Institution)
SRM Nagar, Kattankulathur – 603 203

DEPARTMENT OF INFORMATION TECHNOLOGY

QUESTION BANK

I SEMESTER – M. Tech. - Data Science


1924103– MACHINE LEARNING
Regulation – 2019

Academic Year 2021 – 2022(odd Semester)

Prepared by

Dr.A.R.Revathi, Associate Professor/IT


SRM VALLIAMMAI ENGINEERING COLLEGE
(An Autonomous Institution)
SRM Nagar, Kattankulathur – 603 203
DEPARTMENT OF INFORAMTION TECHNOLOGY
M. Tech- Data Science
QUESTION BANK

SUBJECT : 1924103- MACHINE LEARNING


SEM/YEAR: I / I

UNIT I
UNIT – I : INTRODUCTION 9
Introduction- overview of machine learning- - Generative learning- Gaussian parameter estimation- maximum
likelihood estimation- MAP estimation- Bayesian estimation- bias and variance of estimators- missing and noisy
features- nonparametric density estimation- applications- software tools.
PART – A
Q.No. Questions BT Level Competence
1 Define Machine Learning. BTL 1 Remembering
2 Compare learning vs programming. BTL 2 Understanding
3 Examine the importance of machine learning algorithms. BTL 4 Analyzing
4 Summarize algorithm technique in ML BTL 2 Understanding
5 What are major components involved in each machine algorithm? BTL 1 Remembering
6 List different forms of learning. BTL 1 Remembering
7 Build the machine learning framework. BTL 3 Applying
8 Draw basic learning system model. BTL 3 Applying
9 What are the factors affecting the performance of machine learning algorithm? BTL 1 Remembering
10 Compare generic machine model and discriminate machine learning model. BTL 2 Understanding
11 Discuss major applications of machine learning.
BTL 6 Creating
12 Define Gaussian process.
BTL 2 Understanding
13 Distinguish Joint probability vs conditional probability.
BTL 4 Analyzing
14 Point out/examine supervised learning category and technique.
BTL 4 Analyzing
15 Define maximum likelihood estimation (MLE).
BTL 2 Understanding
16 Explain semi supervised learning?
BTL 5 Evaluating
17 Explain Inductive machine learning?
BTL 5 Evaluating
18 How AI differ from Machine learning? BTL 1 Remembering
19 Discuss the difference between training set and testing set.
BTL 6 Creating
20 List the application software tools.
BTL 1 Remembering

PART – B
Q.No. Questions Marks BT Level Competence
1 What is machine learning? Discuss about learning and machine learning. 13
Choose various types of machine learning. BTL 1 Remembering

2 Demonstrate the supervised learning structure. 13 BTL 2 Understanding


3 Explain briefly about unsupervised learning structure? 13 BTL 2 Understanding
4 Examine in detail about machine learning process with an example. 13 BTL 4 Analyzing
5 Explain various learning techniques involved in supervised learning? 13 BTL 5 Evaluating
6 Explain various learning techniques involved in unsupervised learning? 13 BTL 5 Evaluating
7 How to develop /frame inductive learning and summarize the machine 13
learning process? BTL 1 Remembering

8 Explain generative machine learning model and analyze how it differ from 13
discriminative machine learning model? BTL 4 Analyzing

9 What is Gaussian process? And explain in detail of Gaussian parameter 13


estimates with suitable examples. BTL 1 Remembering

10 Develop procedure in parameter estimation for Bayesian parameter 13


estimation. BTL 3 Applying

11 Discuss about MLE and how it derive from MAP and extreme estimator. 13
BTL 6 Creating
12 Summarize the advantages and disadvantages of Gaussian process 13
BTL 2 Understanding
13 Explain list of application software tools used in machine learning and 13
describe each tools in details. BTL 1 Remembering

14 Short note on a) Bias and variance 6


b) Bias and variance trade off 7 BTL 2 Understanding

PART – C
Q.No. Questions Marks BT Level Competence
1 Find the covariance and correlation coefficient of data X={1,2,3,4,5} 15
BTL 6 Creating
and Y={1,4,9,16,25)
2 Define SVD. Explain the procedure for decomposition matrix with 15
suitable example BTL 5 Evaluating

3 Discuss the difference and tradeoff between two factors of bias and 15
variance. How these factor play a role in machine learning models. BTL 6 Creating
Explain.

4 Explain what are the various metrics used to evaluate the machine 15
learning model performance BTL 5 Evaluating

UNIT II
Neural networks- perceptron- multilayer perceptron’s - back propagation nonlinear regression - multiclass
discrimination- training procedures-localized network structure- dimensionality reduction interpretation
PART – A
Q.No. Questions BT Level Competence
1 Compare biological neuron and artificial neuron BTL 5 Evaluating
2 Draw the structure of artificial single neuron based on biological neuron BTL 3 Applying
3 List out major parts of biological neuron BTL 1 Remembering
4 Discuss about types of artificial neural networks BTL 6 Creating
5 What is the drawback of McCulloch & Pitts mathematical model of an artificial
BTL 1 Remembering
neuron?
6 Define perceptron BTL 2 Understanding
7 Draw the simple perceptron model. BTL 3 Applying
8 Identify the parameters in a perceptron network and its significance BTL 3 Applying
9 Why XOR problem could not be solved by simple perceptron? BTL 1 Remembering
10 Define multilayer perceptron BTL 2 Understanding
11 How a multilayer perceptron solve XOR problem? BTL 4 Analyzing
12 What are activation function? BTL 1 Remembering
13 List some linear and nonlinear activation function BTL 1 Remembering
14 Apply the formula for sigmoidal function BTL 3 Applying
15 Examine the delta learning rule BTL 4 Analyzing
16 How to estimate error in the output layer? BTL 2 Understanding
17 How to evaluate the update weight in back propagation? BTL 2 Understanding
18 When to use the regression? BTL 1 Remembering
19 What is dimensionality reduction? BTL 1 Remembering
20 Justify the necessity for dimensionality reduction in the context of machine
BTL 5 Evaluating
learning.

PART – B
Q.No. Questions Marks BT Level Competence
1 Explain biological Neuron. 13 BTL 2 Understanding
2 Explain simple model of an Artificial Neuron and its functions. 13 BTL 2 Understanding
3 Construct and explain Artificial Neural network structure. 13 BTL 3 Applying
4 Determine activation function and list few activation function with 13
BTL 5 Evaluating
description.
5 Summarize and Explain various types of artificial neural network. 13 BTL 2 Understanding
6 a) Explain application of ANN and list the challenges of ANN. 6
7 BTL 1 Remembering
b)list Advantages and disadvantages of ANN.
7 Develop simple perception model and explain learning theory. 13 BTL 3 Applying
8 Analyze the XOR is not linearly separable? Justify how it can be solved. 13 BTL 4 Analyzing
9 List the factors affecting MLP performance and explain each. 13 BTL 1 Remembering
10 What is the importance of MLP? Explain learning in MLP. 13 BTL 1 Remembering
11 a) Compare Linear Vs nonlinear regression. 6
7 BTL 4 Analyzing
b) Explain back propagation nonlinear regression.

12 Explain Back propagation network. 13 BTL 2 Understanding


13 Discuss the steps involved in Back propagation algorithm. 13 BTL 6 Creating
14 How dimensionality reduction is important in NN? Justify. 13 BTL 1 Remembering

PART – C
Q.No. Questions Marks BT Level Competence
1 Design single layer perceptron with two iteration. Consider the
perceptron having with the initial weights w1=0.5, w2 = 0, learning rate
α=0.2 and bias ᶿ =0.4 for AND Boolean function. The activation function 15
BTL 6 Creating
is the Step function f(x) which gives the output either 0 or 1. If value of
f(x) is greater than or equal to 0, it outputs 1 or else it outputs 0.

2 Consider a perception to represent the Boolean function And with the


initial weights w1=0.3, w2 = -0.2, learning rate α=0.2 and bias ᶿ =0.4 as
shown in Figure. The activation function used here is the Step function
f(x) which gives the output value as binary. i.e., 0 or 1. If value of f(x) is
greater than or equal to 0, it outputs 1 or else it outputs 0. Design a
perceptron that performs the Boolean function AND & update the BTL 5 Evaluating
weights until the Boolean function gives the desired output

15
3 Perform a feedforward operation in a Multi-Layer Perception and
conclude the result. This given MLP consists of an Input layer, one
Hidden layer and an Output layer. The input layer has 4 neurons, the
hidden layer has 2 neurons and the output layer has a single neuron and
the Learning rate is 0.8 .

X1 X2 X3 X4 ODesired
1 1 0 1 1

15
BTL 5 Evaluating

4. Design Back propagation using Multi-Layer Perception which has three


layers like the input layer has 4 neurons, the hidden layer has 2 neurons
and the output layer has a single neuron. Train the MLP by updating the 15
BTL 6 Creating
weights and biases in the network. Learning rate: =0.8. Refer Q.no : 3
for the structure of MLP and their weights

UNIT III
UNIT – III : SUPERVISED LEARNING 9
Decision Trees: ID3, Classification and Regression Trees, Regression: Linear Regression, Multiple Linear
Regression, Logistic Regression, Perceptron, Support vector machines - Linear and Non-Linear, Kernel
Functions, K-Nearest Neighbours.
PART – A
Q.No. Questions BT Level Competence
1 Define entropy BTL 2 Understanding
2 Relate entropy and information gain BTL 2 Understanding
3 Define regression BTL 2 Understanding
4 What is the role of regression model in exploratory data analysis? BTL 1 Remembering
5 What is CART? BTL 1 Remembering
6 How does CART solve the regression problems? BTL 4 Analyzing
7 Compare regression and Estimations BTL 5 Evaluating
8 Compare classification and regression models BTL 5 Evaluating
9 What is the principle of ordinary least square in linear regression? BTL 1 Remembering
10 Compare linear regression model and logistic regression model BTL 4 Analyzing
11 Identify pros and cons of regression models. BTL 3 Applying
12 Develop the concepts of K- Nearest Neighbours. BTL 3 Applying
13 What are benefits of K- NN algorithm? BTL 1 Remembering
14 Identify the disadvantage of K- NN algorithm BTL 3 Applying
15 Discuss how data normalization / standardization is required in K-NN? BTL 6 Creating
16 List out the advantages of SVM BTL 2 Understanding
17 What do you understand by similarity based learning? BTL 1 Remembering
18 Discuss instance based learning vs model based learning. BTL 6 Creating
19 How does the structure of decision tree help in classifying a data instance? BTL 4 Analyzing
20 What are the different metrics used in deciding the splitting attribute? BTL 1 Remembering

PART – B
Q.No. Questions Marks BT Level Competence
1 Build the structure of a decision tree. 13 BTL 3 Applying
2 Explain logistic regression with suitable example 13 BTL 2 Understanding
3 Discuss about linear regression and derive the Individual error & 13
BTL 6 Creating
Minimization functions.
4 Write short notes on (i) Regression and Correlation 6
7 BTL 1 Remembering
(ii) Limitation of Regression model
5 Explain the difference between linear and logistics regression with 13
BTL 5 Evaluating
example.
6 What are the Metrics used to validate the result of regression and 13
BTL 1 Remembering
explain each.
7 How to construct Regression tree and write procedure to construct 13
BTL 1 Remembering
regression tree with example.
8 Explain CART (Classification & Regression tree) algorithm with 13
BTL 2 Understanding
example.
9 a) Compare Instance based learning and Model based learning. 6
7 BTL 4 Analyzing
b) List example of Instance-based learning algorithm.
10 What is the role of kernels? Classify the different type of Kernel. 13 BTL 4 Analyzing
11 List the advantages of SVM and how optimal Hyperplane differ from 13
Hyper plane BTL 1 Remembering

12 Explain Soft margin support vector machine. 13 BTL 2 Understanding


13 Explain weighted K-nearest Neighbor algorithm. 13 BTL 2 Understanding
14 Develop the following (i) Kernel based non-linear classifier. 6
7 BTL 3 Applying
(ii) Support Vector Regression.

PART – C
Q.No. Questions Marks BT Level Competence
1 How to construct ID3 and derive the procedure to construct a decision 15
BTL 6 Creating
tree using ID3
2 Explain SVM classifier with suitable example 15 BTL 5 Evaluating
3 Consider the training dataset given in the following table. Use Weighted
k-NN and determine the class. Test instance (7.6, 60, 8) and K=3.

S.No. CGPA Assessment Project Result


Submitted
1 9.2 85 8 Pass
2 8 80 7 Pass 15 BTL 6 Creating
3 8.5 81 8 Pass
4 6 45 5 Fail
5 6.5 50 4 Fail
6 8.2 72 7 Pass
7 5.8 38 5 Fail
8 8.9 91 9 Pass
4 Consider the training dataset shown in the Table and construct a decision
tree using CART algorithm BTL 6 Creating
S.no CG Inter Practical Communi Job
PA active Knowledge cation Offer
Skills
1 >=9 Yes Very good Good Yes
2 >=8 No Good Moderate Yes 15
3 >=9 No Average Poor No
4 <8 Yes Average Good No
5 >=8 Yes Good Moderate Yes
6 >=9 Yes Good Moderate Yes
7 <8 Yes Good Poor No
8 >=9 No Very good Good Yes
9 >=8 Yes Good Good Yes
10 >=8 Yes Average Good Yes

UNIT IV
UNIT – IV : UNSUPERVISED LEARNING 9
Introduction to clustering, Hierarchical: AGNES, DIANA, Partitional: K-means clustering, K-Mode Clustering,
Expectation Maximization, Gaussian Mixture Models. Bayesian Learning, Bayes Optimal Classifier, Naıve Bayes
Classifier, Bayesian Belief Networks
PART – A
Q.No. Questions BT Level Competence
1 Define clustering. BTL 2 Understanding
2 List out the disadvantages of clustering schemes. BTL 4 Analyzing
3 Distinguish between classification and Clustering. BTL 4 Analyzing
4 List out the applications of clustering algorithm. BTL 1 Remembering
5 Identify the challenges of clustering algorithm. BTL 3 Applying
6 Estimate the problems associated with clustering large data. BTL 5 Evaluating
7 What is k in k-means algorithm? How it is selected? BTL 1 Remembering
8 What is meant by probabilistic based learning? BTL 1 Remembering
9 Define Objective probability BTL 2 Understanding
10 Define subjective probability BTL 2 Understanding
11 Discuss Bayesian probability BTL 6 Creating
12 Explain conditional probability BTL 2 Understanding
13 Explain joint probability BTL 2 Understanding
14 Compare probabilistic model and deterministic model BTL 5 Evaluating
15 Develop the procedure for agglomerative algorithm. BTL 3 Applying
16 Discuss Bayesian network BTL 6 Creating
17 What is belief measure? BTL 1 Remembering
18 State Bayes theorem BTL 1 Remembering
19 What is meant by Bayesian belief network (BBN)? BTL 1 Remembering
20 Choose type of inference performed in BBN? BTL 3 Applying

PART – B
Q.No. Questions Marks BT Level Competence
1 Explain the concepts of clustering approaches. How it differ from 13
2 Understanding
classification.
2 List the applications of clustering and identify advantages and 13
disadvantages of clustering algorithm. 1 Remembering

3 Explain about Hierarchical clustering algorithm. 13 2 Understanding


4 Develop Mean Shift Clustering algorithm. 13 3 Applying
5 Recall the steps involved in Partitional clustering algorithm. 13 1 Remembering
6 Explain about EM algorithm. 13 2 Understanding
7 Write short notes on a) Cohesion & Separation 6
1 Remembering
7
b) Silhouette Co-efficient

8 Discuss the fundamentals of Bayes theorem. 13 4 Analyzing


9 Explain the classification using Bayes Model. 13
4 Analyzing
10 Develop the following a) Bayes Optimal Classifier. 6
7 3 Applying
b) Gibbs Algorithm

11 Explain about Naïve Bayes algorithm for continuous attributes with 13


examples. 4 Analyzing

12 Explain about various Bayesian classifier. 13


5 Evaluating
13 Consider a boy who has a volleyball tournament on the next day, but today 13
he feels sick. It is unusual that there is only a 40% chance he would fall
sick since he is a healthy boy. Now, Find the probability of the boy
participating in the tournament. The boy is very much interested in volley 1 Remembering
ball, so there is a 90% probability that he would participate in tournaments
and 20% that he will fall sick given that he participates in the tournament.

14 Design and discuss how to construct BBN. 13


6 Creating

PART – C
Q.No. Questions Marks BT Level Competence
1 5
a) If the coordinates of the objects are (0,-3) and (5,8) then what is the
Chebyshev distance.
5 BTL 5 Evaluating
b) Discuss MIN algorithm with suitable examples
c) Discuss Quantitative variables evaluation in clustering algorithm
5
2 Compile the single linkage algorithm for the following array points

Objects X Y
15
0 1 4

1 2 8 BTL 6 Creating
2 5 10

3 12 18

4 14 28

3 Cluster the following set of data using k-means algorithm with initial
value of objects 2 and 5 with the coordinate values (4,6) and (12,4) as
initial seeds.

Objec X-coordinate Y-coordinate


ts 15 BTL 5 Evaluating
1 2 4
2 4 6
3 6 8
4 10 4
5 12 4
4 a) Discuss about Bayesian inferences 5
10 BTL 6 Creating
b) Explain Top down inference & Bottom-up reasoning approaches.
UNIT V

UNIT – V : PATTERN RECOGNITION 9


Introduction: Basics of pattern recognition – Design principles of pattern recognition system – Learning and
adaptation – Pattern recognition approaches. Parameter Estimation Techniques, Non-parametric Estimation
Techniques, Linear Discriminant Functions, Dimension Reduction Techniques, Nonmetric Data
PART – A
Q.No. Questions BT Level Competence
1 Define Pattern? BTL 1 Remembering
2 Explain pattern recognition BTL 2 Understanding
3 Compare statistical and structural approaches in pattern recognition BTL 5 Evaluating
4 State Bayes theorem. BTL 1 Remembering
5 Illustrate the random variable. BTL 2 Understanding
6 Examine the parametric estimation method. BTL 4 Analyzing
7 Distinguish between maximum likely hood and Bayes’ method. BTL 4 Analyzing
8 Explain feature extraction. BTL 5 Evaluating
9 List properties of expectation-maximization algorithm. BTL 1 Remembering
10 What is clustering? BTL 1 Remembering
11 List Parametric and Non-Parametric techniques BTL 2 Understanding
12 Develop the F1 score? How would you use it? BTL 3 Applying
13 Apply Precision and Recall to any model? BTL 3 Applying
14 Organize how to Tackle Overfitting and Under fitting? BTL 3 Applying
15 Summarize Ensemble learning. BTL 2 Understanding
16 Explain Correlation and Covariance? BTL 5 Evaluating
17 Build the design pattern recognition system. BTL 3 Applying
18 Discuss on Principal Component Analysis. BTL 6 Creating
19 Discuss on non-parametric methods. BTL 6 Creating
20 Define dimensionality reduction. BTL 1 Remembering

PART – B
Q.No. Questions Marks BT Level Competence
1 Explain general principles of likelihood estimation. 13 BTL 2 Understanding
2 Recall maximum Likelihood estimation with mean and variance. 13 BTL 1 Remembering
3 When the maximum likelihood and Bayes’ method differ? Explain. 13 BTL 3 Applying
4 Explain principle component analysis. 13 BTL 2 Understanding
5 Write an algorithm for expectation-maximization. 13 BTL 1 Remembering
6 Derive the computation of hidden Markov model. 13 BTL 4 Analyzing
7 Estimate density Estimation under non-parametric method. BTL 5 Evaluating
8 Write Short notes on BTL 6 Creating
(a) Parzen-Window Estimation 6
(b) Estimation of Posterior Probability. 7
9 What is Linear Discriminant Function and explain its categories. 13 BTL 1 Remembering
10 Write a Descent Algorithm under Relaxation procedures. 13 BTL 1 Remembering
11 Describe Support Vector Machine. How the vector developed in the BTL 3 Applying
13
training pattern.
12 Discuss SVM for XOR problems. 13 BTL 2 Understanding
13 Examine Pattern Recognition Systems and explain each BTL 4 Analyzing
13
components.
14 Construct a general Pattern Recognition Systems under design BTL 6 Creating
cycle. 13

PART – C
Q.No. Questions Marks BT Level Competence
1 List non-parametric techniques and Explain K-nearest neighbor 15 BTL 4 Analyzing
estimation.
2 Define pattern. Develop the design cycle of pattern recognition 15 BTL 6 Creating
system and explain components involved in PR system.
3 List parametric techniques and explain any one. 15 BTL 4 Analyzing
4 Explain about Dimension reduction and its techniques 15 BTL 5 Evaluating

You might also like