First page is coversheet

No page numbers.
Page number starts from second page – Table of Content
But not Indo-Arabic instead use Roman Letters numbering system (lower is better)
After table of content go with List of Table, List of Figures and List of Abbreviations follow.
Up to this point use roman letters numbering system (i, ii, iii, iv, v…….)

Starting from chapter 01 use Indo-Arabic numbering system till end of chapter 02
Start in a new page and page count starts from here and you cannot have more than 20 pages
Chapter 01 and 02

CHAPTER 01: PROBLEM

1.1 Chapter Overview
1.2 Introduction

1.3 Problem Background / Problem Domain

1.4 Problem Definition

1.4.1 Problem Statement

Explain the problem in nutshell in one sentence

1.5 Motivation

1.6 Existing Work

Citation Summary Limitation Contribution

1.7 Research Gap

What is the Gap: why it is gap with citation to justify the gap

If you are addressing 5 gaps then repeat the same

8. Contribution to Body of Knowledge

1. Contribution to Problem Domain

2. Contribution to Research Domain
If we take the example of “Deep Reinforcement Learning for Autonomous Drone Navigation
in Complex Environments”

Research Domain? Deep Reinforcement Learning

Problem Domain? Autonomous Drone Navigation

Among the above which contribution most important?

If there are no proper contribution to the research domain your marks will be capped at 69
and to score 70 and above you must have contribution to research domain.

How to document the contribution?

1. Contribution to Problem Domain

What is the contribution: Why it is a contribution – justify if possible bring evidence

from Literature

Repeat the above if there are more than one contribution

2. Contribution to Research Domain

What might be contribution and why that might be a contribution

1.9 Research Challenge

What is the challenge in addressing the research gap and fulfilling the contribution that you
have identified? Higher the research challenge better the marks. This is the section where
you are convincing the audience of this research.

What is the challenge? Justify why is it a challenge and if possible, bring the citations from
LR

Repeat the above for each challenges.

Research Questions

Maximum of 3 to 4 research questions

Answer to the research question is the project thesis

Must be specific and avoid vague or broader ones

Also avoid yes or no type of question No “Can a ML model build to predict CKD at early
stage” – no such can type of question because the answer is Yes / No
Instead rephrase it – How can a ML model built to predict CKD at early stage

1.10 Research Aim

The aim of this research is to design, develop and evaluate a machine learning model to
predict CKD at early stage.

In the net paragraph you can further elaborate the research aim

1.11 Research Objectives

List of research objectives and objectives are further breakdown like a check list to achieve
the aim and answer the research question also to achieve the learning outcomes.

Undergraduate LOs

Cyber Security and Forensics Students Please pay attention to:

deploy appropriate tools and technologies to implement software artefacts

that satisfy specified requirements and/or test/evaluate their use in a target
domain;
Research objectives are list of to do task to complete the project (Check List)

Research objectives must be SMART

S – Specific
M – Measurable
A – Achievable
R – Relevance
T – Timebound

Keep the above in mind when writing the Research Objectives

1.12 Chapter Overview

CHAPTER 02 METHODOLOGY
2.1 Chapter Overview
2.2 Research Methodology

Use saunder’s research onion to explain research methodology

Use table:

Layer What is being using Why you are using it

Philosophy The possible philosophies in Computer
Science research is

1. Positivism – you use positivism if There are four types of data:

your research is quantitative – Nominal
everything is a number
Ordinal
2. Interpretivism – qualitative
Interval

3. Pragmatism – they mix Ratio

positivism and interpretivism
These are qualitative in
nature where you wont be
quantifying instead you will
qualitative way of analysis

Mixed methodology
Approach Deductive – you deduce the theory – From theory => hypothesis
theory already exist you check if this ML => collect data => test your
theory can be used to predict / detect hypothesis => accept or
something that is not done before or new reject => knowledge
algorithm has come can we use that to creation
get a better prediction.

Inductive – theory building process

Start from observation =>

identify pattern => come
with theory to explain the
pattern => knowledge
creation
Methodological Mono Method – only one method
choice
Multi method – more than one method
but all from either qualitative methods or More than one method same
 quantitative method paradigm either Positivism
or interpretivism
Mixed method – more than one method
but you are mixing quantitative and
qualitative methods to collect data
Strategy What is the data collection strategy you Data collection strategies
will be using can be

1. Survey
2. Interview
3. Observation
4. Self-evaluation
5. Archival research
6. Brain storming
7. Experiment
Time Horizon Cross section This could be for your
dataset for training and
Data are collection is done on one point testing ML model, This
in time could be for RE or
evaluation
Longitudinal
Two or more different time
frames but from same
subjects (participants from
whom we collect data )
Data Collection Population size
and Analysis Sample size
What is the strategy to
collect data
How to analyze that data

You must use either

interview or questionnaire

2.3 Development Methodology

What is development methodology you will be using?

There are so many developments methodology suitable for research project

1. prototyping
2. One Person Scrum
3. One person any other Agile??

But definitely not waterfall method for research project for that purpose not for any software
development project.

Requirement elicitation methodology

Design methodology

Programming Paradigm

Evaluation (Testing) Methodology

Solution methodology

Example if ML:

1. Dataset
2. Data preprocessing
3. Feature selection and engineering
4. Model selection
5. Training
6. Testing
7. Feedback loop – backpropagation or reinforcement learning

2.4 Project Management Methodology

What is the project management methodology you will be using

Prince 2 or Agile Prince 2

1. Project Scope
1. In scope
2. Out scope
3. Diagram depicting the prototype feature.
2. Schedule
1. Gantt chart
2. Deliverables and date
3. Resource requirements
1. Hardware requirement
2. Software Requirement
3. Data requirements – only if it is data science or you need data
4. Skill requirements – Hard skills – technical skills
4. Risks and mitigation
5 top risk
2.5 Chapter Summary

Indo-Arabic numbering system ends here

List of References

Start the list of references in the new page.

Use different page numbering system from this point onwards (Roman Letters upper case
preferably if you have used lower case in the top)

This is not part of 20 pages.

This must have minimum of 20 references and in that minimum 10 related to existing work in
that minimum 60% from high impact score journals from IEEE, ACM, Science Direct,
Elsevier, Web of Science, etc… 40% from reputed authors (check coauthor or author H
Index) published in the conferences (Conference Proceeding)