Sinhgad Technical Education Society’s
NBN SINHGAD TECHNICAL INSTITUTES CAMPUS, PUNE - 41
Department of Computer Engineering
Preliminary Examination
Subject: MACHINE LEARNING
Class : BE Semester: I / II
Date : Maximum Marks: 70
Time : 2½ Hours Minimum Marks for Passing: 28
Instructions: 1. Draw suitable sketches whenever necessary.
2. Figures to the right indicate full marks.
3. Assume suitable data whenever necessary and mention it clearly.
4. Use of logarithmic tables, slide rules, electronic pocket calculator is allowed.
Marks Level CO
R,U/A,A/E,C
Q1 (a) Explain the following terms with suitable examples. [06] U,E CO3
i) Bias ii) Variance iii) Under fitting and Over fitting
(b) b) Differentiate between Lasso Regression and Ridge A CO3
Regression. [06]
(c) Explain gradient descent algorithm with example. [06] U CO3
OR
Q2 (a) What do you mean by regression? Explain with suitable [06] E,C CO3
example.
(b) Write a short note on [06] E,C CO3
i) MAE ii) RMSE iii) R2
(c) What is gradient descent? Compare batch gradient and U,A CO3
[06]
stochastic gradient descent.
Q3 (a) Explain with example the variant of SVM, the support vector [06] U CO4
regression
(b) What do you mean by ensemble learning? Differentiate [06] U,A CO4
between bagging & boosting.
(c) What are different variants of multi-class classification? U,A CO4
Explain them with suitable example. [06]
OR
Q4 (a) Calculate macro average precision, macro average recall and [06] U,R CO4
macro average F-score for the following given confusion matrix
of multi-class
classification.
(b) Write a short note on : i) Random forest. [06] C,A CO4
ii) Adaboost.
(c) Discuss K-nearest neighbour algorithm with suitable example. U CO4
[06]
Q5 (a) With reference to Clustering explain the issue of “Optimization [06] U,A CO5
of Clusters”.
(b) Compare Hierarchical clustering and K-means clustering. [06] A CO5
(c) Explain how a cluster is formed in the density based clustering [06] U CO5
algorithm.
OR
Q6 (a) How would you choose the number of clusters when designing U,R CO5
a K-Medoid clustering algorithm? [06]
.
(b) Write a short note on out lier analysis with respect to clustering. [06] C,A CO5
(c) Differentiate between K-means and Spectral clustering [06] A CO5
Q7 (a) What are building blocks of neural network, elaborate? [06] U,E CO6
(b) Describe characteristics of back propagation algorithm. [06] U CO6
(c) Write a short note on Recurrent neural n/w & convolutional [06] C,A CO6
neural n/w.
OR
Q8 (a) Explain artificial neural n/w based on perception concept with [06] U,A CO6
diagram.
(b) Describe multi-layer neural n/w. Explain why back propagation U CO6
algorithm is required. [06]
(c) Write short note on with refernce with CNN. C,A CO6
i) Convolution layer ii) Hidden layer [06]
