Download as pdf or txt
Download as pdf or txt
You are on page 1of 4

Course Code : CDT 301 GHXW/MW – 22 / 1635

Fifth Semester B. Tech. ( Computer Science and Engineering /


Data Science ) Examination

MACHINE LEARNING

Time : 3 Hours ] [ Max. Marks : 60

Instructions to Candidates :—
(1) All questions carry marks as indicated against them.
(2) Assume suitable data wherever necessary.
(3) Illustrate your answers wherever necessary with the help of neat sketches.

1. (a) Apply Candidate - Elimination Algorithm on learning task given in Table1 :

ID A B C D E Output
1. A1 B1 C1 D2 E1 No
2. A2 B2 C1 D1 E2 Yes
3. A2 B3 C1 D1 E3 Yes
4. A2 B1 C1 D2 E2 Yes 5(CO1)

(b) Using ID3 algorithm, find the root of Decision Tree for the data given
in Table2 :

ID A B C Target
1. F T F T
2. T F T F
3. T T T F
4. F F T T 5(CO1)

GHXW/MW-22 / 1635 Contd.


2. (a) Give an example (in three dimensional space with coordinate values) in
which k - Nearest Neighbors and Distance - Weighted k - Nearest Neighbors
Algorithms (for k = 3) classify a query point in two different classes.
5(CO1)

(b) A college professor believes that if the grade for internal examination is
high in a class, the grade for external examination will also be high. A
random sample of 7 students in that class was selected, and the data
is given below :

Internal Exam 15 23 18 23 24 22 22

External Exam 49 63 58 60 58 61 60

Apply linear regression on the above dataset, and find the coefficients a and
b. Predict the external marks if internal marks are 24. 5(CO1)

3. (a) Consider a medical diagnosis problem in which there are two alternative
hypotheses : (1) That the patient ; has a - particular form of cancer, and
(2) That the patient does not. The available data is from a particular laboratory
test with two possible outcomes ; positive and negative. We have prior
knowledge that over the entire population of people only .006 have this
disease.
Futhermore, the lab test is only an imperfect indicator of the disease. The
test returns a correct positive result in only 99% of 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. In other cases, the
test returns the opposite result.
Suppose we now observe a new patient for whom the lab test returns
a positive result. Should we diagnose the patient as having cancer or not ?
Suppose the doctor decides to order a second test for the same patient
from the same laboratory, and suppose the second test returns a positive
result as well. What are the posterior probabilities of cancer and not cancer
following these two tests ?
Assume that the two tests are independent. 5(CO2)

GHXW/MW-22 / 1635 2 Contd.


(b) Consider the following Bayesian network. Suppose that the network further
records the following probabilities :

Compute P(D), P(A | C) and P(~D | C). 5(CO2)

4. (a) State the Apriori Property. Using Apriori Algorithm find the final item set
for the following dataset S. Where (min-Sup = 50%, min_conf = 70%).
Generate all association rules and list the strongest rule :
TID Items Purchased
101 Book, Note, Pen
102 Pencil, Note, Eraser
103 Book, Pencil, Note, Eraser
104 Pencil, Eraser 5(CO2)
(b) Apply FP Tree algorithm to find frequent itemset from the given transactional
database. Min_support = 2. Generate the FP tree and frequent itemset :
TID Items
1 P, Q
2 Q, R, S
3 P, R, S, T
4 P, S, T
5 P, Q, R 5(CO2)

GHXW/MW-22 / 1635 3 Contd.


5. (a) Apply K-means clustering algorithm on following data and identify cluster
for each individual. (Assume k = 2) :
Object X Y
Weight pH value
index
A 1 1
B 3 2
C 4 3
D 5 4 4(CO3)
(b) Use following dataset and perform hierarchical clustering using complete-linkage
algorithm. Show dendogram result :
X Y
1 4 4
2 8 4
3 15 8
4 24 4
5 24 12 6(CO3)
6. (a) Find out a linear SVM to solve the following classification problem :
Positive examples : {(–1, 2), (–3, 4), (–4, 4), (–5, 5)}
Negative examples : {(3, –2), (4, –3), (3, –3), (5, –4)} 5(CO4)
(b) Construct a Strong classifier using decision tree stumps (weak classifiers)
to solve the following classification problem :
X1 X2 Class
1 1 negative
2 2 positive
3 3 negative 5(CO4)

D
GHXW/MW-22 / 1635 4 25

You might also like