USN

16LNI322/SCE321/SCN324/SCS31/
SFC254/SIT322/SSE334
Ph.D./M.Sc. (Engg.) by Research Degree Examination, Nov./Dec.2020
Machine Learning Techniques

Time: 3 hrs. Max. Marks: 100

Note: Answer any FIVE full questions, choosing ONE full question from each module.
2. Any revealing of identification, appeal to evaluator and /or equations written eg, 42+8 = 50, will be treated as malpractice.

Module-1
1 a. Describe the steps in designing a learning system. (10 Marks)
b. Write candidate-elimination algorithm. Apply the same on the following training examples:
1 Sky Temp Humidity Wind Water Forecast Enjoy sport
2 Sunny Warm Normal Strong Warm Same Yes
Important Note : 1. On completing your answers, compulsorily draw diagonal cross lines on the remaining blank pages.

3 Sunny Warm High Strong Warm Same Yes

4 Rainy Col High Strong Warm Change No
(10 Marks)

OR
2 a. List any five applications of machine learning and discuss the issues in machine learning.
(10 Marks)
b. What are decision trees? Define the terms entropy and information gain with respect to
decision trees. (05 Marks)
c. Given the following partition [Fig.Q2(c)], compute the value of information gain.

Fig.Q2(c) (05 Marks)

Module-2
3 a. Explain in detail any five appropriate problems for neural network learning. (10 Marks)
b. What is artificial neural network? A single neuron is shown in the Fig.Q3(b):

Fig.Q3(b)

1 of 3
 16LNI322/SCE321/SCN324/SCS31/SFC254/SIT322/SSE334
The node has 3 inputs X = (X1, X2, X3) that receive only binary signals (0/1). The weights
corresponding to the inputs have the following values w1 = 2, w2 = 4, w3 = 1. The
activation function is given as:
1 if v  0
f (v)  
0 otherwise
(i) How many different input patterns this node can receive?
(ii) Calculate what will be the output value ‘y’ of the unit for each of the following input
patterns.
Pattern P1 P2 P3 P4
X1 1 0 1 1
X2 0 1 0 1
X3 0 1 1 1
(10 Marks)
OR
4 a. What is multi-layers neural network? Write and explain the Back propagation algorithm for
feed forward neural network. (10 Marks)
b. Write the prototypical genetic algorithm. Apply the genetic operators on the following bit
strings:
Masks
 single pt crossover:
str1: 11101001000 11111000000
str2: 00001010101  2-pt crossover:
00111110000
 Uniform crossover:
10011010011 (10 Marks)

Module-3
5 a. What is naïve Bayes classifier? Consider the problem of classifying days according to
whether someone will play tennis. The table below provides a set of 14 training examples of
the target concept play tennis, where each day is described by the attributes: outlook,
temperature, humidity and wind. Apply naïve Bayes classifier to classify following instance:
<Outlook = Sunny, Temperature = cool, Humidity = High Wind = Strong>
Day Outlook Temperature Humidity Wind Play Tennis
D1 Sunny Hot High Weak No
D2 Sunny Hot High Strong No
D3 Overcast Hot High Weak Yes
D4 Rain Mild High Weak Yes
D5 Rain Cool Normal Weak Yes
D6 Rain Cool Normal Strong No
D7 Overcast Cool Normal Strong Yes
D8 Sunny Mild High Weak No
D9 Sunny Cool Normal Weak Yes
D10 Rain Mild Normal Weak Yes
D11 Sunny Mild Normal Strong Yes
D12 Overcast Mild High Strong Yes
D13 Overcast Hot Normal Weak Yes
D14 Rain Mild High Strong No
(10 Marks)
b. Explain mistake bound for FIND-S and HALVING algorithms. (10 Marks)
2 of 3
 16LNI322/SCE321/SCN324/SCS31/SFC254/SIT322/SSE334

OR
6 a. Explain sample complexity for infinite hypothesis spaces. (10 Marks)
b. Define the terms supervised and unsupervised learning. A patient takes a lab test for cancer
diagnosis. There are two possible outcomes in this case: + (positive) and  (negative). The
test returns a correct positive result in only 98%. If the cases in which the disease is actually
present and a correct negative result in only 97% of the cases in which the disease is not
present. Furthermore, 0.008 of the entire population have this cancer. Compute the values of
the following:
(i) P (cancer) (ii) P( cancer) (iii) P(+|cancer)
(iv) P(|cancer) (v) P(+|cancer) (vi) P(|cancer) (10 Marks)

Module-4
7 a. Explain K-nearest neighbor algorithm. Discuss problem associated with this algorithm and
solutions to this problem. (10 Marks)
b. What is case based reasoning? List its generic properties. (05 Marks)
c. Describe the issues associated with case based reasoning. (05 Marks)

OR
8 a. What is inductive logic programming? Write and explain FOIL learning algorithm.
(10 Marks)
b. What is sequential covering? Write LEARN-ONE-RULE algorithm for general to specific
beam search. (10 Marks)

Module-5
9 a. What is correctness and completeness of domain theory? (05 Marks)
b. Which technique is used by the KBANN algorithm? Explain the need of KBANN algorithm.
(05 Marks)
c. Explain the capabilities of explanation based learning. (10 Marks)

OR
10 a. Describe three different methods for using prior knowledge to alter the search performed by
inductive methods. (10 Marks)
b. What is reinforcement learning? How it differs from other function approximation tasks?
(10 Marks)

*****

3 of 3