Module’s Assessment Schedule

Notes on
cw1 Research Report
1
Module’s Assessment Schedule

Form of
assessment
Assignment Weighting Submission date
Coursework #1 Research Report 70% Friday 18th November
2022
Coursework #2 Data Ethics
Report
30% Friday 23rd December
2022

2
cw1 is about
cw1 is about ML’s security
Synonyms:
Security / privacy of machine learning
Attacks & defences in ML
Vulnerabilities / threats of ML
3
cw1 is NOT about
cw1 is NOT about ML for security
Synonyms:
AI/ML for cybersecurity
Applying AI/ML for cybersecurity
Cybersecurity AI/ML/analytics
4
Example Topics for cw1

# Working Title Keyword 1 Keyword 2 Keyword 3
(optional)
1. security and privacy of
machine learning
machine learning security privacy
vulnerability
2. attack and defence in
machine learning
machine learning attack defence
vulnerability
3. adversarial machine
learning
adversarial machine
learning
4. security and privacy of
deep learning
deep learning security privacy
vulnerability
5. security and privacy of
reinforcement learning
reinforcement
learning
security privacy
vulnerability
6. privacy-preserving
machine learning
machine learning privacy preserving
7. federated learning federated learning
8. attack and defence in
federated learning
federated learning attack defence
vulnerability
9. algorithmic bias algorithmic bias machine learning

5
How To Conduct Literature Review
GCU library guide
Systematic review
https://www.gcu.ac.uk/currentstudents/essentials/li
brary/research/systematicreview
6
Similar guides
Conduct a literature review
https://new.library.arizona.edu/research/writecite/lit-review
7 easy steps to do a literature review
https://www.concordia.ca/cunews/offices/vprgs/grad
proskills/blogs/2020/06/04/7-easy-steps-to-do-aliterature-review.html
How to Write a Literature Review | Guide, Examples,
& Templates, by Shona McCombes
https://www.scribbr.com/methodology/literaturereview/
7
Taxonomy
Oxford dictionary,
https://www.oxfordlearnersdictionaries.com/
definition/english/taxonomy?q=taxonomy
taxonomy noun
[uncountable] the scientific process of classifying
things (= arranging them into groups)
[countable] a particular system of classifying
things
8
Taxonomy
Classification/categorisation of the things of
concern
The things of concern for cw1 are two sides
Problems/issues of the topic selected/studied;
Solutions/approaches to the problems/issues
The taxonomy for your cw1 would be a mind map
illustrating and describing
the problems/issues of your topic selected/studied;
the corresponding solutions/approaches to the
problems/issues
9
Example of Taxonomy
10
Example of Taxonomy
11
Example of Taxonomy
12
Relationships Between Chapters
There are prerequisite relationships from one to the
next between Chapters 1, 2, 3 & 4.
In Chapter 1, you will delineate a clear scope of the
topic you have selected
By Chapter 2, you will establish a good understanding
of the topic you have studied.
In Chapter 3, you will provide an overview upon the
toolkits & frameworks relating to your topic.
Then, in Chapter 4, you will select one of the toolkits /
frameworks based on the overview of Chapter 3, to
conduct your experiments
13
Kickstart
1. Go to the coursework # 1 folder on GCULearn
2. Pick up one from the example topics
3. Read the indicative readings in the coursework # 1 folder
4. Go to the major databases as stated in week 1’s folder,
search more targeted papers of your topic
5. Read papers and write to cw1 report according to the
template structure as stated in the cw1 specifications
6. Then, write an overview on the relevant software toolkits
of your topic
7. Select one software toolkit, install & set up, and conduct
your experiments
14
Research Questions
By asking even simple questions about your
Topic,
it will actively engage yourself into ways of
studying/exploring into the Topic
Providing answers requires you to develop the
necessary understanding on the Topic
15
Research Questions: an example
Take data/model poisoning attack as an example,
simple questions:
What does it mean by data/model poisoning attack?
Where & how can a data/model poisoning attack occur in
the machine learning process?
What impact will a data/model poisoning attack make
upon the performance of machine learning?
How many types of data/model poisoning attacks are
there?
What are the defence strategies against each type of
data/model poisoning attack
How effective is each defence strategy?
what are the pros & cons?
16
Refine Research Questions
While research is progressed into the Topic,
simple research questions will be refined to be
more essential ones, and more targeted ones
17
Research Questions: some guides
Research question
https://en.wikipedia.org/wiki/Research_question
How to Write a Good Research Question (w/ Examples)
https://blog.wordvice.com/how-to-write-a-hypothesis-orresearch-question/
How to Write a Research Question: Types, Steps, and
Examples
https://research.com/research/how-to-write-aresearch-question
Formulation of Research Question – Stepwise Approach
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6322175/
Narrowing a Topic and Developing a Research Question
https://libraries.indiana.edu/sites/default/files/Develop_a_
Research_Question.pdf
18
Research Problem
Research problem is about your Topic
More specifically, research problem is your
quantification of the Topic
19
Research Problem: some guides
What is a Research Problem? by Jamie
Walker
https://www.researchprospect.com/what-is-a-researchproblem/
Problem statement
https://en.wikipedia.org/wiki/Problem_statement
how to define a research problem with example | step by step
guide
https://www.youtube.com/watch?v=QB6FdwLDiI&ab_channel=Educationalhub
How to Define a Research Problem | Ideas & Examples, by Shona
McCombes
https://www.scribbr.com/research-process/researchproblem/
How to Write a Problem Statement | Guide & Examples, by Shona
McCombes
https://www.scribbr.com/research-process/problemstatement/
20