Total No. of Questions : 8] SEAT No.

8
23
PA-913 [Total No. of Pages : 3

ic-
[5927]-343

tat
1s
B.E. (Computer)

5:1
02 91
MACHINE LEARNING

3:4
0
(2019 Pattern) (Semester - VII) (410242)

31
2/0 13
0
1/2
Time : 2½ Hours] [Max. Marks : 70
.23 GP

Instructions to the candidates:

E

1) Solve Q.1 or Q.2, Q.3 or Q.4, Q.5 or Q6, Q.7 or Q.8.

81

8
C

23
2) Neat diagrams must be drawn wherever necessary.

ic-
3) Figures to the right indicate full marks.
16

tat
8.2

4) Make suitable assumption whenever necessary.

1s
.24

5:1
91
49

3:4
Q1) a) Explain in brief techniques to reduce under fitting and over fitting. [6]
30
31

b) Find the Equation of linear Regression line using following data : [6]
01
02

X Y
1/2
GP
2/0

1 3
CE
81

8
2 4

23
.23

3 5
ic-
16

tat
4 7
8.2

1s
.24

5:1

c) Write short note on : [6]

91
49

3:4

i) MAE
30
31

ii) RMSE
01
02

iii) R 2
1/2
GP
2/0

OR
CE
81

Q2) a) Explain in brief lasso and Ridge Regression. [6]

.23

b) What is Bias and variance trade off for machine learning model? [6]
16
8.2

c) Write short note on Evaluation metrics. [6]

.24
49

P.T.O.
Q3) a) Explain in brief methods used for Evaluating classification models.

8
[5]

23
ic-
b) Consider the following data to predict the student pass or fail using the

tat
K-Nearest Neighbor Algorithm (KNN) for the values physics = 6 marks,

1s
Chemistry = 8 marks with number of Neighbors K = 3. [6]

5:1
02 91
Physics (marks) Chemistry (marks) Results

3:4
0
4 3 Fail

31
2/0 13
6 7 Pass
0
1/2
.23 GP

7 8 Pass
5 5 Fail
E
81

8
C

23
8 8 Pass

ic-
16

tat
c) Write short note on Ensemble learing methods : [6]
8.2

1s
i) Simple
.24

5:1
91
ii) Advanced
49

3:4
30
31

OR
01
02

Q4) a) Explain Random forest Algorithm with example. [5]

1/2
GP

b) Write short note on importance of confusion matrix. [6]

2/0
CE

c) Define following terms with reference to SVM. [6]

81

8
23
.23

i) Separating hyperplane
ic-
16

tat
ii) Margin
8.2

1s
.24

5:1
91
49

Q5) a) Explain Density Based clustering with refence to DBSCAN, OPTICS

3:4
30

and DENCLUE. [6]

31
01

b) What is K mean clustering? Explain with example. [6]

02
1/2
GP

c) Write short note on following Hierarchical clustering method : [6]

2/0

i) Agglomerative
CE
81

ii) Dendogram
.23
16

OR
8.2
.24
49

[5927]-343 2
Q6) a) What is LOF? Explain it with it's advantages and disadvantages. [6]

8
23
b) Explain Graph Based clustering. [6]

ic-
c) Define following terms : [6]

tat
1s
i) Elbow method

5:1
02 91
ii) Extrinsic and Intrinsic method

3:4
0
31
2/0 13
Q7) a) Explain ANN with it's Architecture. [5]
0
1/2
.23 GP

b) Obtain the output of Neuron Y for the Network shown in following

fig. Using activation function as : [6]
E
81

8
C

23
i) Binary sigmoidal

ic-
ii) Bipolar sigmoidal
16

tat
8.2

1s
.24

5:1
91
49

3:4
30
31
01
02
1/2
GP
2/0
CE
81

8
23
c) Write short note on Back propagation network. [6]
.23

OR ic-
16

tat
8.2

1s

Q8) a) Explain in brief types of ANN based on layers. [5]

.24

5:1

b) What is Recurrent Neural Network? Explain with suitable example.

91
49

3:4

[6]
30
31

c) Write short note on with refernce with CNN. [6]

01
02

i) Convolution layer
1/2
GP
2/0

ii) Hidden layer

CE
81
.23


16
8.2
.24
49

[5927]-343 3
Total No. of Questions : 8] SEAT No. :

8
23
P546 [Total No. of Pages : 2
[6004]-481

ic-
tat
B.E. (Computer Engineering)

5s
MACHINE LEARNING

7:2
(2019 Pattern) (Semester - VII) (410242)

02 91
4:3
Time : 2½ Hours] [Max. Marks : 70

0
31
Instructions to the candidates:
2/0 13
1) Attempt Q.1 or Q.2, Q.3 or Q.4, Q.5 or Q.6 and Q.7 or Q.8.
0
2) Figures to the right indicate full marks.
6/2
.23 GP

3) Neat diagrams must be drawn wherever necessary.

4) Assume suitable data, if necessary.
E
82

8
C

23
ic-
Q1) a) Explain the following terms with suitable examples. [6]
16

tat
i) Bias
8.2

5s
ii) Variance
.24

7:2
91
iii) Under fitting and Over fitting
49

4:3
30

b) Differentiate between Lasso Regression and Ridge Regression. [6]

31
01

c) Explain gradient descent algorithm with example. [6]

02
6/2
GP

OR
2/0

Q2) a) What do you mean by regression? Explain with suitable example. [6]
CE
82

38
b) Write a short note on : [6]

c-2
.23

i) MAE
i
16

tat
ii) RMSE
8.2

5s

iii) R2
.24

7:2
91
49

c) What is gradient descent? Compare batch gradient and stochastic gradient

4:3
30

descent. [6]
31
01
02
6/2

Q3) a) Explain with example the variant of SVM, the support vector regression.
GP

[5]
2/0

b) What do you mean by ensemble learning?Differentiate between bagging

CE
82

& boosting. [6]

.23

c) What are different variants of multi-class classification? Explain them

16

with suitable example. [6]

8.2

OR
.24

P.T.O.
49
Q4) a) Calculate macro average precision, macro average recall and macro

8
23
average F-score for the following given confusion matrix of multi-class

ic-
classification. [6]

tat
Predictions 

5s
A B C D

7:2
A 100 80 10 10

02 91
4:3
B 0 9 0 1

0
Actual values 

31
2/0 13 C 0 1 8 1
D 0 1 0 9
0
6/2
.23 GP

b) Write a short note on : [6]

i) Random forest.
E
82

8
ii) Adaboost.
C

23
c) Discuss K-nearest neighbour algorithm with suitable example. [5]

ic-
16

tat
8.2

5s
Q5) a) With reference to Clustering explain the issue of “Optimization of Clusters”.
.24

7:2
[6]
91
49

b) Compare Hierarchical clustering and K-means clustering. [6]

4:3
30

c) Explain how a cluster is formed in the density based clustering algorithm.

31

[6]
01
02

OR
6/2
GP

Q6) a) How would you choose the number of clusters when designing a K-
2/0

Medoid clustering algorithm? [6]

CE
82

38
b) Write a short note on out lier analysis with respect to clustering. [6]

c-2
.23

c) Differentiate between K-means and Spectral clustering. [6]

i
16

tat
8.2

5s

Q7) a) What are building blocks of neural network, elaborate? [5]

.24

7:2

b) Describe characteristics of back propagation algorithm. [6]

91
49

4:3

c) Write a short note on Recurrent neural n/w & convolutional neural n/w.
30

[6]
31

OR
01
02

Q8) a) Explain artificial neural n/w based on perception concept with diagram.
6/2
GP

[6]
2/0

b) Describe multi-layer neural n/w. Explain why back propagation algorithm

CE
82

is required. [6]
.23

c) Discuss any two activation functions with example. [5]

16


8.2
.24

[6004]-481 2
49
Total No. of Questions : 8] SEAT No. :

8
23
P-6552 [Total No. of Pages : 2

ic-
[6181]-102

tat
5s
B.E. (Computer Engineering)

7:5
02 91
MACHINE LEARNING

3:3
0
(2019 Pattern) (Semester - VII) (410242)

31
2/1 13
0
Time : 2½ Hours] [Max. Marks : 70
2/2
.23 GP

Instructions to the candidates:

1) Answer Q1 or Q2, Q3 or Q4, Q5 or Q6, Q7 or Q8.
E
80

8
C

23
2) Neat diagrams must be drawn wherever necessary.

ic-
3) Figures to the right side indicate full marks.
16

tat
4) Assume suitable data if necessary.
8.2

5s
.24

7:5
Q1) a) Differentiate between overfitting and underfitting. [6]
91
49

3:3
b) The table below shows the number of grams of carbohydrates, X and
30
31

the number of Calories, Y of six different foods. Find linear regression

equation for this dataset. [8]
01
02
2/2

Carbohydrates (X) 8 9.5 10 6 7 4

GP
2/1

Calories (Y) 12 138 147 88 108 62

CE

Also find the value of Y for X = 12

80

8
23
c) Explain Bias Variance Trade off. [4]
.23

OR ic-
16

tat
8.2

Q2) a) What is Linear Regression? Explain the concept of Ridge regression.

5s

[9]
.24

7:5
91

b) Explain the following Evaluation Metrics : [9]

49

3:3
30

i) MAE
31

ii) RMSE
01
02

iii) R2
2/2
GP
2/1
CE

Q3) a) Differentiate between bagging and boosting. [4]

80

b) What is ensemble learning? Explain the concept of Random Forest

.23

ensemble learning. [9]

16

c) What is the relation between precision and recall? Explain with an

8.2

example. [4]
.24
49

P.T.O.
 OR

8
23
Q4) a) What is K-fold cross-validation? In K-fold cross-validation, comment

ic-
on the following situations [9]

tat
i) When the value of K is too large

5s
7:5
ii) When the value of K is too small.

02 91
3:3
How do you decide the value of k in k-fold cross-validation?

0
31
b) 2/1 13
Explain i) Accuracy, ii) Precision, iii) Recall, and iv) F-Score [8]
0
2/2
.23 GP

Q5) a) Explain K-Means clustering in detail with a suitable example. [8]

E
80

8
C

23
b) What is outlier analysis? How is Local Outlier Factor detected? [5]

ic-
c) Explain Spectral Cluster in galgorithm. [5]
16

tat
8.2

5s
OR
.24

7:5
Q6) a) Explain Hierarchical and Density-based Clustering approaches. [9]
91
49

3:3
b) Write short note on : [9]
30
31

i) Optimization of clusters
01
02
2/2

ii) K-Medoids
GP
2/1

iii) Evaluation metrics

CE
80

8
23
.23

Q7) a) Write a note on Single Layer Neural Network. ic-[4]

16

tat
8.2

5s

b) Explain Radial Basis Function networks in detail. [8]

.24

7:5

c) Explain Recurrent Neural Networks and its applications in brief. [5]

91
49

3:3
30

OR
31
01

Q8) a) Explain the concept of Back Propagation in ANN with example. [8]
02
2/2
GP

b) What is Functional Link Artificial Neural Network (FLANN)? Explain

2/1

its merits over other ANNs. [5]

CE
80

c) What is Activation Function? Explain with a suitable example. [4]

.23
16
8.2


.24
49

[6181]-102 2
Total No. of Questions : 8] SEAT No. :

8
23
PB-2244 [Total No. of Pages : 2

ic-
tat
[6263]-82

1s
1:5
B.E. (Computer Engineering)

02 91
3:3
MACHINE LEARNING

0
41
4/0 13
(2019 Pattern) (Semester - VII) (410242)
0
5/2
Time : 2½ Hours] [Max. Marks : 70
.23 GP

Instructions to the candidates :

E
81

1) Answer Q.1 or Q.2, Q.3 or Q.4, Q.5 or Q.6, Q.7 or Q.8.

8
C

23
2) Figures to the right side indicate full marks.

ic-
3) Draw neat diagram wherever necessary.
16

tat
4) Assume suitable data, if necessary.
8.2

1s
.24

1:5
91
Q1) a) Define different regression models. [6]
49

3:3
30

b) What are different techniques to reduce under fitting? [6]

41
01
02

c) With following data of shows company expenditure. [6]

5/2
GP

x(month) 1 2 3 4 5
4/0

y(expenditure) 12 19 29 37 45
CE
81

8
23
using regression model predict expenditure of 6th month.
.23

OR ic-
16

tat
8.2

1s

Q2) a) What is R2 measure of evaluation? [6]

.24

1:5

b) What do you mean by least square method? Explain least square method
91
49

3:3

in the context of linear regression. [6]

30
41

c) Write a short note on stochastic qradient descent algorithms. [6]

01
02
5/2
GP

Q3) a) Why ensemble learning is used for ML? [5]

4/0
CE

b) What are advantages and disadvantages of K-NN? [6]

81
.23

c) What are different distance metrics used in k-NN? [6]

16

OR
8.2

P.T.O.
.24
49
Q4) a) What is multiclass classification? Explain the variants of multiclass

8
23
classification. [5]

ic-
b) Explain kernel methods which are suitable for SVM. [6]

tat
1s
c) What are different techniques used for outlier handling? [6]

1:5
02 91
3:3
Q5) a) Why K-medoid is used? Explain k-medoid algorithm. [5]

0
41
b) 4/0 13
Why density based clustering is used? Explain any one. [6]
0
5/2
c) What is outlier analysis? [6]
.23 GP

OR
E
81

8
Q6) a) What is isolation factor model? [5]
C

23
ic-
b) Explain k means algorithm. [6]
16

tat
c) Explain Hierarchical clustering with example [6]
8.2

1s
.24

1:5
91
49

3:3
Q7) a) What is Multilayer perceptron? Describe with diagram. [6]
30
41

b) What are different activation function used is NN? [6]

01
02

c) Explain Convolution Neural Network. (CNN) with suitable example. [6]

5/2
GP

OR
4/0
CE

Q8) a) Explain building blocks of RBF networks. [6]

81

8
23
.23

b) What is Personalized recommendation? What is content based

recommendation? [6] ic-
16

tat
8.2

1s

c) Explain Recurrent Neural Networks with as example. [6]

.24

1:5
91
49

3:3


30
41
01
02
5/2
GP
4/0
CE
81
.23
16
8.2
.24

[6263]-82 2
49