SPPU Information Technology (Semester 7)
Machine Learning
May 2017
Total marks: --
Total time: --
(1) Assume appropriate data and state your reasons
(2) Marks are given to the right of every question
(3) Draw neat diagrams wherever necessary

Solve any one question from Q 1(a,b) & Q 2. (a, b)
1(a) write mathematical form of the following:
i) Classitication.
ii) Class probability estimation. iii)Regression.
Which one out of these three is more precise? Which one leads to overfitting?
4 M
1(b) Prove with an example FP=Neg-TN.
4 M

2(a) Write output code matrix for one-versus-one symmetric case. Assume three classes.
4 M
2(b) Justify use of Machine Learning to solve following task:?Prediction of sale value of a car based on the locality of the property?.
4 M

Solve any one question from Q 3(a,b) & Q 4. (a, b)
3(a) Explain VC dimention.
4 M
3(b) Explain Kernel methods for non-linearity.
4 M

4(a) What is Machine Learning? Explain any one application where Machine Learning can be used.
4 M
4(b) Explain Support Vector Machine.
4 M

Solve any one question from Q 5(a,b) & Q 6. (a, b)
5(a) Final all 3- item itemsets from this set with minimum support=2.
Trans_id Itemlist
T1 {K, A, D, B}
T2 {D, A, C, E, B}
T3 {C, A, B, E}
T4 {B, A, D}
9 M
5(b) Write K-means Algorithm.
9 M

6(a) Explain silhouettes.
9 M
6(b) Discuss various distance measures.
9 M

Solve any one question from Q 7(a,b) & Q 8. (a, b)
7(a) Write a note on compression based models.
8 M
7(b) Explain Naive Bayes Classification Algorithm
8 M

8(a) Difine the terms:
i) Bernoulli distribution.
ii) Binomial distribution
iii) Multinomial distribution
iv) Gaussion distribution.
8 M
8(b) Explain Discriminative learning.
8 M

Solve any one question from Q 9(a,b) & Q 10. (a, b)
9(a) Explain on-line learning.
8 M
9(b) Explain multi task learning.
8 M

10(a) Explain the concept of penalty and award in reinforcement learning.
8 M
10(b) Explain ensemble learning.
8 M

More question papers from Machine Learning