You are on page 1of 3

University of Central Punjab

Faculty of Information Technology

Assignment 4

Course Title: Introduction to Machine Learning


Course Code: SEAD-3453

Name: _______________________________

Registration Number: ___________________

Section: ______________________________

Instructions:
1. Attempt all questions.
2. Write your answer showing all steps required to perform the task.
3. Assignment should be submitted on A4 sheets or Assignment sheets only.
Violation will result to deduction of 1 mark from the scored marks.
4. Each student will have attached this front page with his/her assignment.
Violation will result to deduction of 1 mark from the scored marks.
5. Due Date for Assignment is Jan 21, 2023.
6. Late submission will result in 15% deduction in marks.
7. No request for late submissions will be considered after two working days
of the deadline.
8. No request to review assignment will be considered after 7 working days
of the deadline.
Problem 1 (CLO 02) Marks: 30

A. Suppose that the data mining task is to cluster points into three clusters using K-Means.
The points are (2,10), (2,5), (8,4), (5,8), (7,5), (6,4), (1,2), 𝑎𝑛𝑑(4,9). The distance is
Euclidean distance.
Suppose initially we assign 𝐴 = (2,10), 𝐵 = (5,8), and 𝐶 = (1,2) as the center of each
cluster, respectively.

B. Apply the DBSCAN algorithm with similarity threshold of 0.8 to the given data points and
MinPts≥2 (Minimum required points in a cluster) what are core, border and noise (outliers)
in the set of points in the table below.
Sample POINTS

S1 (5,7)

S2 (8,4)

S3 (3,3)

S4 (4,4)

S5 (3,7)

S6 (6,7)

S7 (6,1)

S8 (5,5)

C. For the given dataset compute the clusters using a single link technique. Use Euclidean
distance and draw the Dendrogram.
Sample POINTS

S1 (0.4,0.53)

S2 (0.22,0.38)

S3 (0.35,0.32)

S4 (0.26,0.19)

S5 (0.08,0.41)

S6 (0.45,0.30)
Problem 2 (CLO 02) Marks: 10

For given dataset, apply Apriori algorithm to discover strong association rules among tags.

Assume that min_support=40% and min_confidence=70%.

Generate association rules from the frequent items. Calculate the confidence of each rule and
identify all the strong association rules.

Transaction ID Items Purchased

1 Bread, Butter, Milk

2 Bread, Butter

3 Beer, Cookies, Diapers

4 Milk, Diapers, Bread, Butter

5 Beer, Diapers

Problem 3 (CLO 02) Marks: 10

Assume that the neurons have a sigmoid activation function, perform and compute a forward pass
and backward pass on the network. Assume that the actual output of 𝑦 is 0.5 and learning rate is
1. Perform another forward pass.

W13 = 0.1 Y3
X1=0.35 H3
W35 = 0.3

W14 = 0.4
O5

W23 = 0.8

W45 = 0.9
X2=0.9 H4
W24 = 0.6
Y4

You might also like