BITS WILP Machine Learning Mid-Sem Exam 2016-H2

BITS WILP Machine Learning Mid-Sem Exam 2016-H2

Birla Institute of Technology & Science, Pilani

Work-Integrated Learning Programmes Division

First Semester 2016-2017

Mid-Semester Test  (EC-2 Regular)

Course No.                  : IS ZC464 

Course Title                 : MACHINE LEARNING 

Nature of Exam           : Closed Book

Weightage                    : 35%

Duration                      : 2 Hours 

Date of Exam              : 25/09/2016    (FN)

No. of pages: 1; No. of questions: 3

Note:

1.       Please follow all the Instructions to Candidates given on the cover page of the answer book.

2.       All parts of a question should be answered consecutively. Each answer should start from a fresh page. 

3.       Assumptions made if any, should be stated clearly at the beginning of your answer.

Q1. Answer the following questions                                                                              [4 ´ 3 = 12]

a) Describe the meaning of 'best hypothesis' in the context of function approximation in machine learning .

b) What is Bayes' theorem? How is it significant in machine learning?

c) Explain MAP technique used in learning? Give example.

d) Explain the role of 'error' in prediction of target value given the test data. Also explain the role of training in prediction. 

Q2. Answer the following questions                                                                   [7 + 7 =14]

a) The chances of children in primary schools in villages dropping (D) their studies are high due to various reasons.  The major factors are lack of basic needs of children at home (N) and  lack of infrastructure (I) such as school building and availability of teachers who can teach well and motivate student. The statistics collected as the joint probabilities are given in the following table.

	I		~ I
	N	~N	N	~N
D	0.098	0.022	0.06	0.02
~D	0.018	0.062	0.32	0.4

Use Bayes' theorem to compute the posterior probability P(D | N) using the given joint probabilities. Explain all steps of calculation. [Note: A calculation without the correct expression will not be given credit.]

b) Consider a linear model of the form          

                                                  

where    is the vector of parameters and is represented as W.  The function  are the basis functions. Explain the significance of the parameters W in linear regression.  Comment on the parameters W in the context of approximating data using a straight line.

Q3. Answer the following questions                                                                    [5 + 4 = 9]

a) What do you understand by entropy? Calculate the entropy for the following data.

Symbols->	A	B	C	D	E
Probability->	0.20	0.15	0.10	0.35	0.20

b) What is the significance of attribute selection in decision tree based learning? Explain with an appropriate example.

NOTE: IF YOU HAVE ANY QUESTION PAPERS FOR POSTING ONLINE, PLEASE SHARE AT 'ashishjainblogger@gmail.com'