UNIVERSITY EXAMINATIONS: 2017/2018
EXAMINATION FOR THE DEGREE OF MASTER OF SCIENCE IN DATA ANALYICS
MDA 5202 MACHINE LEARNING
DATE: DECEMBER 2018 TIME: 2 HOURS
INSTRUCTIONS: Answer Question One & ANY OTHER TWO questions.
a) Discuss the three categories of machine learning 6 Marks
b) Discuss the importance of machine learning in the world today 3 Marks
c) Using an example explain how regression is applied in machine learning 3 Marks
d) Explain the characteristics of support vector machines 4 Marks
e) With an aid of a pseudocode or an algorithm describe the steps followed in K-Means
technique 4 Marks
a) Discuss how Bayesian theorem can be used to implement the following systems
i. Medical diagnosis systems 2 Marks
ii. Manufacturing system diagnosis 2 Marks
iii. Computer systems diagnosis 3 Marks
iv. Information retrieval 3 Marks
b) Examine the tools available in python to facilitate machine learning 3 Marks
c) Assume that we observe the weather over a ten-day period. In particular, we observe the
1. The sun shines on the first day.
2. It rains on day 5.
3. It rains on day 7.
4. The sun shines on day 10.
What is the probability of sun on day 6? Derive the equation 2 Marks
a) Analyse any four reasons for advocating for neural networks to solve problems
b) What is the goal of Principle Component Analysis (PCA)? 2 Marks
c) Using an example differentiate between clustering and classification 2 Marks
d) Discuss the applicability of manifold learning 3 Marks
Suppose that we want to solve the problem of finding out what a
good car is by using genetic algorithms. Suppose further that the solution to the
problem can be represented by a decision tree as follows:
a) What is the appropriate chromosome design for the given problem? 3 Marks
b) Which Genetic Algorithm parameters need to be defined? 3 Marks
c) What would be the suitable values of those parameters for the given problem? 3 Marks
d) What is the result of applying a single round of the prototypical Genetic Algorithm?
Explain your answer in a clear and compact manner by providing the pseudo code of the
algorithm. 6 Marks