You are on page 1of 1

Sample Exam from the previous batch

1. Is it possible to use support vector machine (SVM) for prediction? (Yes/No) If the answer is no
why? else how it’s used? (4 pt)

2. Given the following data


Point (p) 0.5 3.0 4.2 4.4 4.6 4.9 5.2 5.4 5.6 7.0 8.0
Class A A B B B A B B A B B
Classify the data point p= 4.75 based on 3- and 5-nearest neighbors’ algorithm and
justify the classification decision. (4 pt)

3. Given the following data for Naive Bayes (NB) model, we want to determine the
probability of the players will play if weather for humidity High? (4 pt)
Humidity High High High High Normal Normal Normal High Normal Normal Normal High
Play No No Yes Yes Yes No Yes No Yes Yes Yes Yes

4. Consider the current conflict that happens in different part of the country in Ethiopian. There are
many reasons for the conflict. Among these reason, Fake News disseminated over the social media
is one. Considering the case of Fake News, you have been selected as machine learning expert to
design a machine learning system that can detect a fake news from the real one and report for the
respective office. Based on the above information answer the following three question (A, B and
C). assuming you have unlimited access to technology such as telecom access is provided from
Ethio-telecom
4A. What type of data is required and how you collect to design such a ML system? (2.5pt)
4B. Which ML algorithm is appropriate to create model? Why? (2.5pt)
4C. Explain how you implement a ML model that you have selected in the above answer. Give
a brief description of the system and explain how to deploy the model to solve the stated
problems. (4pt)
5. Consider the given 9 data points shown in table of two-dimensional data of x and z. Assumption
use P1 and P6 are selected as initial centroid for cluster 1 and 2, respectively.
Subject P1 P2 P3 P4 P5 P6 P7 P8 P9
A 1 1.5 3 5 3.5 4.5 3.5 1.5 2.5
B 1 2 4 7 5 5 4.5 3 4

5a. Apply K-means clustering algorithm to group the datasets in two cluster. (5 pt)
5b. When do we terminate the iteration 4a of K-means clustering. (3 pt)

You might also like