Scribd
Upload
27 views2 pages

MTech (DS) Sem-II Data Mining and Predictive Analytics - Out

Uploaded by

Harsh roy
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
27 views2 pages

MTech (DS) Sem-II Data Mining and Predictive Analytics - Out

Uploaded by

Harsh roy
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

ENROLLMENT NUMBER

MARKING SCHEME
Amity University Haryana
M.Tech. (DS)
Second Semester End Term Examinations – April-May, 2019
Course Title: Data Mining and Predictive Analytics

Course Code: DSE4202


Exam Dept. Ref.: Q039UY
Time: 3 Hours Maximum Marks: 70
Note: This question paper has three sections: Section A, Section B and Section C.
Attempt (i) Any FIVE from Sec.-A.
(ii) Any TWO from Sec.-B.
(iii) Section-C is compulsory.

Section –A (Attempt any five questions) 30 Marks


(Each question carries 6 marks)

1. In real-world data, tuples with missing values for some attributes are a common
occurrence.Describe methods for handling this problem.
Answer: Name of methods -2 Marks, Explanation- 4 Marks.

2. Describe the steps that are required in a data mining process and explain the importance
of each step.
Answer: Steps – 3 Marks, Explanation of importance of steps – 3 Marks.

3. Managers want to know by next week whether deployment will take place.
Therefore,analysts meet to discuss how useful and accurate their model is. Explain which
phase in the CRISP-DM process is represented in the stated scenario.
Answer: Name of steps – 3 Marks, Explanation – 3 Marks.

4. Suppose that our target variable is continuous numeric. Can we apply decision trees
directly to classify it? How can we work around this?
Answer: Answer andWorking Mechanism – 6 Marks.

5. Discuss association. How we can measure the usefulness of association rules?


Answer: Definition – 2 Marks, Justification – 4 Marks.

6. Describe some of the similarities between Kohonen networks and the neural networks.
Answer: 4 Similarities – 6 marks.

Section – B (Attempt any two questions) 20 Marks


(Each question carries 10 marks)

7. Explain Decision Tree Induction.Describe basic algorithm for inducing a decision tree
from training tuples?
Answer: Definition decision tree – 3 marks, Algorithm – 3 Marks.
Page 1 of 2
8. Outliers are often discarded as noise. However, one person’s garbage could be another’s
treasure. For example, exceptions in credit card transactions can help us detect the
fraudulent use of credit cards. Taking fraudulence detection as an example, propose two
methodsthat can be used to detect outliers and discuss which one is more reliable.
Answer: Name of methods-4 Marks, Explanation and name of outlier – 6 Marks.

9. Discuss Cluster Analysis. What are some typical applications of clustering? What are
some typical requirements of clustering in data mining?
Answer: Definition- 3 Marks, Application – 3 Marks, Requirement – 4 Marks.

Section – C (Compulsory) 20 Marks

10. a) Consider the following database consisting of 6 transactions:


S no Transaction ID Items
1 101 A,C,D
2 102 B,C,D,E
3 103 B,D,E
4 104 B,C,E
5 105 A,B,D
6 106 B,C,D

Suppose minimum support count is fixed as 50% and minimum confidence required is
70%. Find out the frequent item set using the apriori algorithm. (10)
Answer: Demonstration of algorithm – 10 Marks

b) Using the data in below table, find the k-nearest neighbour for record 5 and record 10
using k=3. (10)

Record Age Marital Income Risk


1 22 Single $46,156.98 Bad loss
2 33 Married $24,188.10 Bad loss
3 28 other $28,787.34 Bad loss
4 51 other $23,886.72 Bad loss
5 25 Single $47,281.44 Bad loss
6 39 Single $33,994.90 Good Risk
7 54 Single $28,716.50 Good Risk
8 55 Married $49,186.75 Good Risk
9 50 Married $46,726.50 Good Risk
10 66 Married $36,120.34 Good Risk

Answer: For Each record – 5 Mark.

Page 2 of 2

You might also like