UCSP 0010 – Research Methodology

MODULE 2
PROBLEM FORMULATION

Professor Dr Naomie Salim

Faculty of Computing
Universiti Teknologi Malaysia

The image part

with www.utm.my innovative ● entrepreneurial ● global 1
 Outline
• What is a research problem?
• Steps in formulating a research problem
• Writing about research problems

The image part

with www.utm.my innovative ● entrepreneurial ● global 2
 Definition of a Problem
• A problem is defined as a situation
where two conditions exist:
– There is a gap between an existing state
and a desired state
– The solution for closing the gap has not
been determined
• Therefore we need to know as much as
possible about the topic before we can
formulate a researchable problem
The image part
with www.utm.my innovative ● entrepreneurial ● global 3
 What is a Research Problem?

A research problem is basically a “gap” between “what

is “ and “what ought to be”.
Tips :
i. Find out why the information is being sought
ii. Determine whether the information already exist
iii. Determine whether the question really can/should be
answered
iv. Use exploratory research to define background of the
problem
v. Situation analysis
vi. The iceberg principle
vii. Determine relevant variables.
The image part 4
with INSPIRING CREATIVE AND INNOVATIVE
www.utm.my innovative ● MINDS
entrepreneurial ● global 4
 Problem Formulation
• Problem solving has two phases:
– What is the problem that you want to solve
• Defining a problem
– How can the problem be solved
• Discover a solution to a problem (Develop a
plan to solve a problem)
• Formulation means translating and transforming the
selected research problem into a scientifically answerable
research question
• The problem formulation should provide a clear
statement of the purpose and scope of study

The image part

with www.utm.my innovative ● entrepreneurial ● global 5
 Problem Formulation
A situation, person
or things that
needs attention Problem
and needs to be
dealt with or solved

Problem
Formulation

To develop all the

details of a plan for Formulation
solving problem

The image part

with www.utm.my innovative ● entrepreneurial ● global 6
 What is a good problem for
you
A good problem for an individual has the following
characteristics:
• You find the problem interesting
• You feel the problem is something you can tackle
in a reasonable amount of time.
• You care about the problem enough to expend
resources on in terms of time, effort, money, etc.
• You think you might be able to make a contribution
to solving the problem or part of it.

The image part

with www.utm.my innovative ● entrepreneurial ● global 7
 Topic Identification
– Can you finish it on time?
Does It Suit You? How Fast Can You
• Student’s Background Start?
• Student’s Interests • Equipment/Hardware
• Student’s Preparation • Data Availability
• Student’s Capabilities • Software/Simulator

Supportive Literature Support?

Environment? • Is it well-published? In
• Supervisor’s Knowledge reputable venues?
• Supervisor’s Research • Is it a growing research
Experience area?
• Senior Students • Can the data and
methodology be obtained
• Available Networks
from literature ?
Understood by student?

The image part

with www.utm.my innovative ● entrepreneurial ● global 8
 “Man approaches the unattainable truth
through a succession of errors.”

-Aldous Huxley-

The image part

with www.utm.my innovative ● entrepreneurial ● global 9
 Finding A Good Research
Problem in Computer
Science
It is about studying an Idea: your Idea.
Novelty of the Idea.
Research is a study of new ideas in the field where the
research belongs to.
Significance for the Community.
One of the most important questions of research is to study
what idea is actually needed for the community “today”.
Contribution from the Researcher.
An amount of efforts made by a researcher to study the
idea.
But before...

The image part

with www.utm.my innovative ● entrepreneurial ● global 10
 But before...
... understanding Novelty and Significance you have to know the
state-of-the-art of the Scientific Community.
How to be up-to-date?
1. Read recent journal articles, and conference papers. Almost
all of them has “History”, “Introduction” and “Future work” parts.
(they correspond to “Past”, “Current” and “Possible Future” of
the research.)
2. Talk to colleagues and scientific advisers :) (they may
suggest ideas and explain the field development, without
studying).
3. See the business tendency and technology levels (news
from industry).
4. Read the views of the future (Sometimes knowledgeable
people publish their visions of the future).
The image part
with www.utm.my innovative ● entrepreneurial ● global 11
 HOW TO SELECT A PROBLEM?

• Personal practical experience.

• Previous or prior work on subject.
• Own curiosity or own great mind
• Critical Study of the literature search.
• New technology.
• Interaction with other researchers in the
community eg correspondence, seminars,
conferences, colloquium, etc.

The image part

with INSPIRING CREATIVE AND INNOVATIVE
www.utm.my innovative ● MINDS
entrepreneurial ● global 12
 Problem Formulation in Research
Process: The Scientific Method

• Step 1: State the Problem (Problem

Formulation)
• Step 2: Background & Literature study about
the problem. (Literature Review)
• Step 3: Form a Hypothesis (Data Collection)
• Step 4: Do experiment that test the
hypothesis. (Analysis & Findings)
• Step 5: Draw a conclusion. (Derive
Conclusions)
The image part
with INSPIRING CREATIVE AND INNOVATIVE
www.utm.my innovative ● MINDS
entrepreneurial ● global 13
 Steps in
Problem Formulation
Process
• Step 1: Select a topic
• Step 2: Review literature
• Step 3: Develop Problem
Statement
• Step 4: Formulate Research
questions
• Step 5: Formulate objectives
• Step 6: Define scope
• Step 7: Double check.

The image part

with www.utm.my innovative ● entrepreneurial ● global 14
 Problem Identification
Process
• Identify Domain Area (Broad Topic)
– eg. Cryptography, Watermarking, Biometric Recognition,
Image Segmentation
• Make a conceptual map by identifying related topics
(Narrow Topic)
– Divide the domain area into progressively smaller sub-area
until one reaches a subject of interest
– Tools: mindmap, fishbone
• Which topic of interests you/wish to explore
• Identify “real” problem area in the sub-area
– What is the most important and possible relative to scope
and scale of study?
The image part
with www.utm.my innovative ● entrepreneurial ● global 15
 Research Problem

General
Background/Problem
Statement of
Problem /
Analysis
Research
Questions/
Finding

The image part

with www.utm.my innovative ● entrepreneurial ● global 16
 Will the result be significant?

• The questions include:

üWill the result advance knowledge?

üWill the research have some value?
üWill the results be of interest to others?

The image part 17

with INSPIRING CREATIVE AND INNOVATIVE
www.utm.my innovative ● MINDS
entrepreneurial ● global 17
 Problem tree analysis
• Analysis of researchable issue to gain insights about its possible
cause-effect relationships.
• Helps in identifying the critical areas where an intervention would
provide a solution to the problem of concern
Impacts No Protection of IP Questionable Integrity

Effect Intelligent plagiarism cannot be detected

Problem No way to detect beyond cut-and paste plagiarism

Causes Plagiarist change words Change order of words

The image part

with www.utm.my innovative ● entrepreneurial ● global 18
 Change Problem Tree to
Objective Tree
Impacts Protection of IP Preserve Integrity

Effect Avoid Intelligent plagiarism

Aims Develop ways to detect beyond cut-and paste plagiarism

Objectives Semantic-based detection Predicate-based detection

The image part

with www.utm.my innovative ● entrepreneurial ● global 19
 Problem Tree – Keep asking
Why?
(SA Prathapar, Research Methodology Slide, 2012)
Unskilled Labor Inefficient Irrigation
Low Labor
Productivity

Unsuitable Climate Inefficient Water harvesting

Water Scarcity Lack of crop varieties

adapted to climate
Food Insecurity

Unsuitable Crops Farming Patterns do not

Return nutrients

Low Land Farmers can’t afford fertilizers

Productivity

Farmers unaware of
Poor Soil best practices
12/2/22
The image part 20
with www.utm.my innovative ● entrepreneurial ● global 20
 Example - Identifying Gaps
in Information Retrieval
Research
• Other data that can be used to enhance?
– Eg. Explicit vs implicit, multimedia
• Other ways to represent data?
– Eg. Graph? Passages?
• Challenge assumptions and rules.
– Eg.: crisp vs fuzzy? Deterministic vs probabilistics?
• Other external knowledge bases or sources to
enhance?
• Combination? Optimization? Weightage?
• Adapting ideas form other fields? Eg. Diversity
analysis, Cross-structural theory, Game theory, etc.
The image part
with www.utm.my innovative ● entrepreneurial ● global 21
 Horizontal Analysis

Output:
Decisions?
Impact?

The image part

with www.utm.my innovative ● entrepreneurial ● global 22
 Horizontal Analysis

Input: Output:
Data? Decisions?
Impact?

The image part

with www.utm.my innovative ● entrepreneurial ● global 23
 Horizontal Analysis

Input: Processes? Output:

Data? Decisions?
Impact?
The
image
part
with
relatio
nship
ID
rId2
was
not
found
in the
file.

The image part

with www.utm.my innovative ● entrepreneurial ● global 24
 Horizontal Analysis

Input: Processes? Output:

Data? Decisions?
Impact?
The
image
part
with
relatio
nship
ID
rId2
was
not
found
in the
file.

The image part

with www.utm.my innovative ● entrepreneurial ● global 25
 Example of Horizontal
Analysis
Input: Processes? Crude Palm
• Historical Oil Stock
Stock Values Market
• Other Prediction:
The
Commodities Turning
image
part
• Weather withPoints?
relatio
• News ID Values?
nship

rId2
was
not
found
in the
file.

The image part

with www.utm.my innovative ● entrepreneurial ● global 26
 Ways to identify solution
• Root cause analysis
• Breaking rules and assumptions
• Proxy benchmarking
• Technology analysis
• Outcome analysis
• Cost and duration analysis
• Inversion
The image part
with www.utm.my innovative ● entrepreneurial ● global 27
 Other ways to find a
legitimate Problem
• Supervisor suggestions
• Suggestions of others
• Replicate Research with modifications
• Apply an existing technique to new
domain
• Address Contradiction and Ambiguity
• Challenge Findings
The image part
with www.utm.my innovative ● entrepreneurial ● global 28
 Requirement for Problem
Identification
• Inquisitive and imaginative mind
• Questioning attitude
– Is the problem/topic significant enough?
– Is it feasible (practical/possible for me to do it)?
– Is it clear (unambiguous)?

The image part

with www.utm.my innovative ● entrepreneurial ● global 29
 Selecting and Defining a
Problem
• Identify and state the problem in specific
form
• Identify the variables in the problem
situation and define them adequately
• Generate tentative guesses (hypothesis)
about the relation of the variables or write
explicitly the questions for which answers
are sought
• Evaluate the problem for its researchability
The image part
with www.utm.my innovative ● entrepreneurial ● global 30
 Criteria for Selecting A Problem

1. Interest: Commit yourself to a research.

2. Size: Manageable & doable.

3. Economical: Time and money.

4. Researcher’s Capabilities & Limitations.

5. Uniqueness: Do not duplicate; similar but differentiated

by method, design or sample or perform different
statistical analyses.

The image part

with INSPIRING CREATIVE AND INNOVATIVE
www.utm.my innovative ● MINDS
entrepreneurial ● global 31
 Criteria for Selecting A Problem

6. Start with theories/set of ideas :- easier to start. The

outcome might be a critique to the theory, suggestions
how it can be modified or extended
7. Potential & sufficient outcomes:- valuable results.
Even if you plan to develop something in the end which
might not be achievable the knowledge developed in
the process still has contribution within the indicated
period (3++ years?)
8. Not bias, safe & ethical:- be objective as much as
possible, not causing discomfort & harm to
anyone/anything (social, emotional, physical).

The image part

with INSPIRING CREATIVE AND INNOVATIVE
www.utm.my innovative ● MINDS
entrepreneurial ● global 32
 A research aims to :

i. To clarify an existing theory

ii. To clarify contradictory findings

iii. To correct a faulty methodology

iv. To correct the inadequate or unsuitable use of

statistical techniques

v. To reconcile conflicting opinions

vi. To solve existing practical problems.

The image part
with INSPIRING CREATIVE AND INNOVATIVE
www.utm.my innovative ● MINDS
entrepreneurial ● global 33
 Is the problem researchable?

The questions include:

üHas the problem been specified?
üIs the problem amenable to research?
üIs the problem too large?
üHow is the availability of the data?
üAm I capable of solving the problem?

The image part

with INSPIRING CREATIVE AND INNOVATIVE
www.utm.my innovative ● MINDS
entrepreneurial ● global 34
 Characteristics of Good
Research Problem
• Interesting : keeps the researcher interested in it through the
research process
• Researchable :can be investigated
• Significant : contributes to the improvement and understanding
of computer science theory and practice
• Manageable : fits the level of researcher’s level of research skill,
needed resources (eg. availability of data) and time restrictions
• Ethical : does not embarrass/harm society/participants
• Level of expertise: Make sure that you have adequate level of
expertise for the task you are proposing since you need to do
the work yourself.

The image part

with www.utm.my innovative ● entrepreneurial ● global 35
 Chapter 1 of Thesis
• General Overview/Introduction
• Background of the Problem/Research Background
– To give scholarly background & rationale for the
investigation
– What is the specific problem and how has people solved it
• Problem Statement
– To state in general & specific way The image
part with

• Research Questions relationshi

p ID rId2
was not
• Aim and Objectives (should be measurable) found in
the file.

• Scope (what is covered?)

• Importance/Significance of Research
– Refers to the rationale for the study & its relationship to
theory, knowledge or practice
• Thesis Organization
The image part
with www.utm.my innovative ● entrepreneurial ● global 36
 Problem Formulation Process
vs Thesis Content (Chapter 1)
Problem Formulation Process Chapter 1 Thesis Content
Exploring the topic 1.0 General Overview

Reviewing the literature 1.1 Background of the Problem/Research

Background
Developing a problem statement 1.2 Problem Statement

Formulate research question Research Questions

Formulate research objective 1.3 Aim and Objective

Define research scope 1.4 Scope

1.5 Importance/Significance of Research

1.6 Thesis Organization
The image part
with www.utm.my innovative ● entrepreneurial ● global 37
 To State the Problem

• It takes two forms:

üProblem Statement
eg. This study is designed to measure the effect of the
introduction of information security awareness course
has on the computer science students.
üResearch Question
eg. What effect has the introduction of information
security awareness course had on the perception of the
computer science students?

April 2009
The image part 38
with INSPIRING CREATIVE AND INNOVATIVE
www.utm.my innovative ● MINDS
entrepreneurial ● global 38
 Developing Problem
Statement

The image part

with www.utm.my innovative ● entrepreneurial ● global 39
 Problem Statement
• When a problem has been formulated, it can be represented
as statement of the problem
• Problem statement can be written in one sentence or can be
few paragraphs long extending to more than a page
• It is a specific declaration that summarizes the point of
view of your research.
Th

• It should also justify the problem ID

The image part

with www.utm.my innovative ● entrepreneurial ● global 40
 Steps in Step 1: Contextualize the problem
Writing The problem statement should frame your
Problem research problem in its particular context and
give some background on what is already
Statements known about it.
Practical research problems
For practical research, focus on the concrete
details of the situation:
• Where and when does the problem arise?
• Who does the problem affect?
• What attempts have been made to solve
the problem?

The image part

with www.utm.my innovative ● entrepreneurial ● global 41
 Example

Voter turnout in region X has been decreasing steadily over the past ten
years, in contrast to other areas of the country. According to surveys
conducted by organization Y, turnout is lowest among under-25s and
people on low incomes. There have been some effective attempts at
engaging these groups in other regions, and in the last two elections
parties A and B increased their campaigning efforts in region X, but these
interventions have yet to have any significant effect on turnout.

The image part

with www.utm.my innovative ● entrepreneurial ● global 42
 Steps in Theoretical research problems
Writing For theoretical research, think about the
Problem scientific, social, geographical and/or
historical background:
Statements • What is already known about the problem?
• Is the problem limited to a certain time
period or geographical area?
• How has the problem been defined and
debated in the scholarly literature?

The image part

with www.utm.my innovative ● entrepreneurial ● global 43
 Example

In the past ten years, the “gig economy” has become an increasingly
important segment of the labour market. Under-30s are more likely to
engage in freelance, contracted or zero-hour work arrangements instead
of traditional full-time jobs. Research on the reasons for and
consequences of this shift has focused on objective measures of income,
working hours and employment conditions, but there has been little work
exploring young people’s subjective experiences of the gig economy.

The image part

with www.utm.my innovative ● entrepreneurial ● global 44
 Step 2: Show why it matters
• The problem statement should also address the
relevance of the research: why is it important that the
problem is solved?
• This doesn’t mean you have to do something
groundbreaking or world-changing. It’s more
important that the problem is researchable, feasible,
and clearly addresses a relevant issue in your field.

The image part

with www.utm.my innovative ● entrepreneurial ● global 45
 Practical research problems

• Practical research is directly relevant to a specific

problem that affects an organization, institution,
social group, or society more broadly. To make it
clear why your research problem matters, you can
ask yourself:
– What will happen if the problem is not solved?
– Who will feel the consequences?
– Does the problem have wider relevance (e.g. are similar
issues found in other contexts)?

The image part

with www.utm.my innovative ● entrepreneurial ● global 46
 Example
• Low voter turnout has been shown to have negative
associations with social cohesion and civic
engagement, and is becoming an area of increasing
concern in many European democracies. When
specific groups of citizens lack political
representation, they are likely to become more
excluded over time, leading to an erosion of trust in
democratic institutions. Addressing this problem will
have practical benefits for region X and contribute to
understanding of this widespread phenomenon.

The image part

with www.utm.my innovative ● entrepreneurial ● global 47
 Theoretical research
problems
• Sometimes theoretical issues have clear practical
consequences, but sometimes their relevance is less
immediately obvious. To identify why the problem
matters, ask:
– How will resolving the problem advance understanding of the
topic?
– What benefits will it have for future research?
– Does the problem have direct or indirect consequences for
society?

The image part

with www.utm.my innovative ● entrepreneurial ● global 48
 Example
• In the literature on the gig economy, these new forms
of employment are sometimes characterized as a
flexible active choice and sometimes as an
exploitative last resort. To gain a fuller understanding
of why young people engage in the gig economy, in-
depth qualitative research is required. Focusing on
workers’ experiences can help develop more robust
theories of flexibility and precarity in contemporary
employment, as well as potentially informing future
policy objectives.

The image part

with www.utm.my innovative ● entrepreneurial ● global 49
 Step 3: Set your aims and
objectives
• Finally, the problem statement should
frame how you intend to address the
problem. Your goal should not be to find
a conclusive solution, but to seek out
the reasons behind the problem and
propose more effective approaches to
tackling or understanding it.

The image part

with www.utm.my innovative ● entrepreneurial ● global 50
 The aim is the overall
purpose of your research. It
is generally written in the
infinitive form:
• The aim of this study is to determine…
• This project aims to explore…
• I aim to investigate…

The image part

with www.utm.my innovative ● entrepreneurial ● global 51
 The objectives are the
concrete steps you will take
to achieve the aim:
• Qualitative methods will be used
to identify…
• I will use surveys to collect…
• Using statistical analysis, the research
will measure…

The image part

with www.utm.my innovative ● entrepreneurial ● global 52
 Practical research aims and
objectives
• The aim of this research is to
investigate effective engagement
strategies to increase voter turnout in
region X. It will identify the most
significant factors in non-voting through
surveys and interviews, and conduct
experiments to measure the
effectiveness of different strategies.
The image part
with www.utm.my innovative ● entrepreneurial ● global 53
 Theoretical research aims
and objectives
• This project aims to better understand young
people’s experiences in the gig economy.
Qualitative methods will be used to gain in-
depth insight into the motivations and
perceptions of under-30s engaged in
freelance and zero-hour work across various
industries. This data will be contextualized
with a review of recent literature on the gig
economy and statistical analysis of
demographic changes in the workforce.
The image part
with www.utm.my innovative ● entrepreneurial ● global 54
 Guidelines on Problem
Presentation

• Write opening sentence that stimulates interest as

well as conveys an issue to which a broad
readership can relate.
• Specify the problem or issue leading to the study.
• Indicate why the problem is important.
• Focus the problem statement on the key concept
being tested or explored.
• Refrain from using quotes in the lead sentence.
• Consider numeric information for impact.
• Consider short sentences for impact.

The image part

with INSPIRING CREATIVE AND INNOVATIVE
www.utm.my innovative ● MINDS
entrepreneurial ● global 55
 Problem Statement

• Common Mistakes :-
– the problem to investigate is not stated clearly enough
– the scope is too big
• For example:-
– Finding risk assessment methodology for cloud computing, but
there are smaller steps (small research) to achieve the goal.
• Before can proceed to the research question, the researcher should
conduct literature review to extend the knowledge.

The image part 56

with www.utm.my innovative ● entrepreneurial ● global 56
 Example Problem Statement
• The on-line recognition of control chart patterns (CCPs) may encounter
a continuum of situations between stable processes (random patterns)
to fully unstable processes. Generally, the actual onset of any process
deterioration is unknown. Attempting to recognise at too early a stage
could result in unnecessary recognition of stable processes, high false
alarms and wrong classification. On the other hand, delayed
recognition would be undesirable for preventive action and could be too
risky for potentially catastrophic processes. The existing works have
not adequately addressed the recognition of such CCPs, that is, while
they are developing. They deal mainly with automated recognition of
fully developed patterns. The limited few online CCPR schemes either
unnecessarily recognised stable processes, had poor performance or
are too sketchily described. It is desirable to achieve timely and
accurate recognition for unstable CCPs. Besides, stable processes
should be left running as long as possible and maintain minimum false
alarm rates. Therefore, this study addresses the need for such timely
on-line recognition of developing control chart patterns.
The image part
with www.utm.my innovative ● entrepreneurial ● global 57
 Problem Statement
Example: Statement
It is crucial that control chart
patterns be recognised as they are
developing in a timely and
accurate manner, as waiting for
process deterioration to develop
fully could be too late for
preventive purposes or may even
be catastrophic.
The image part
with www.utm.my innovative ● entrepreneurial ● global 58
 Evaluating the problem
statement
There is an associated checklist for creating a problem
statement.
• Research statement written clearly.
• Problem stated in grammatically complete sentences.
• Problem has clearly stated limitations.
• Statement has potential for leading to important results.
• The statement will lead to the analysis of data.
• The problem is focused enough to lead to an answer with
reasonable effort.
• Problem has been reviewed by someone else who provided
feedback.

The image part

with www.utm.my innovative ● entrepreneurial ● global 59
 Formulating Research
Questions

The image part

with www.utm.my innovative ● entrepreneurial ● global 60
 Research Question
• A Research Question is a statement
that identifies the phenomenon to be
studied
• It is a fuel that drives the scientific
process, guides the research and is the
foundation of any research process.
• A research question is a clear, focused,
concise, complex and arguable question
around which any research is centered.
The image part
with www.utm.my innovative ● entrepreneurial ● global 61
 The Research Question

• Characteristics
- State the main concepts
- Is neutral (debateable)
- Clear and specific
• The question should
- Define the research
- Guide your inquiries
- Frame your arguments
- Be likely to produce your “contributions”

The image part

with www.utm.my innovative ● entrepreneurial ● global 62
 Research Question

• A research question must identify

– The variables under study
– The population being studied
– The testability of the question

The image part

with www.utm.my innovative ● entrepreneurial ● global 63
 Thinking about the question
• What is the problem you are attempting to
address?
• What is the unsolved problem that your research
will attempt to resolve?
- What?
- Why?
- Where?
- When?
- Who?
- How?
The image part
with www.utm.my innovative ● entrepreneurial ● global 64
 Example Proto-Questions
• Is there an algorithm that can solve X?
- Can something be done at all?
• How can this X be improved…?
- Can something be made better?
• Why does X work?
- Why does this give the right answer?
• What is the explanation for the phenomenon
demonstrated by X?
- What are the theoretical underpinnings of this
• Can we apply the technique of Y to X to get Z?
- Can we nail a number of things together and get something
new?

The image part

with www.utm.my innovative ● entrepreneurial ● global 65
 Is the Question Reasonable?
• What is the context of the question?
• Is the question significant?
• What is everyone else doing?
• Is there a point of attack on the problem?
• Do I like the question?
- Am I curious about it enough to pursue it?
• Can it be done in the length of time I am
willing to spend on it?

The image part

with www.utm.my innovative ● entrepreneurial ● global 66
 Research Questions
• Consider the following questions about a new
or existing design or technique :
– Is it viable?
– Is it as good as or better than current practice?
– Which of several design alternatives is best?
– What are its performance limits and capabilities?
– What are its strengths and weaknesses?
– Does it work well for novices, for experts?
– How much practice is required to become
proficient?
The image part
with www.utm.my innovative ● entrepreneurial ● global 67
 Benefits of Getting the
Questions Right at the Start
• Improves clarity of the problem
• Facilitates subsequent steps of the research
process
• Helps select appropriate literature for review
• Anticipates issues, resource needs
• Improves efficiency, minimizes wasted
efforts
The image part
with www.utm.my innovative ● entrepreneurial ● global 68
 Types of Research
• Existence:
Questions(1/2)
– designed to systematically rule out rival explanations.
– For example, can neonates perceive colour?
– Do older adults suffer from general slowing?
• Description and classification:
– Usually call for more than simple description. Answers tend to require statements of
the uniqueness of the description to sub class.
– What are the characteristics of depression?
– Is it variable or invariable?
• Composition:
– Call for analysis or breakdown of whole into its component parts.
– What are the principle parts of memory? What are the stages of sleep?
• Relationship:
– Complex questions about relationships among several variables.
– Is there any association between memory and Intelligent Quotient (IQ)?

• and does not under other conditions?

The image part

with www.utm.my innovative ● entrepreneurial ● global 69
 Types of Research
Questions (2/2)
• Descriptive-Comparative:
– Expected that researcher will ensure that all is same between groups except issue in
question
– . e.g., are women more aggressive than men?
• Causality:
– Does X cause, lead to or prevent changes in Y? Usually this can’t be done in
exploratory research. Such question also indicates which type of research methods we
have to use.
– Does Alcohol intoxication lead to liver failure?
• Causality-Comparative:
– Effects of X are compared with a rival treatment Z and not simply absence of treatment.
– For example, Are aerobic exercises better than problem solving exercises in improving
the cognitive performance of children?
• Causality-Comparative Interaction:
– Does X cause more changes in Y than Z under certain conditions and does not under
other conditions?

The image part

with www.utm.my innovative ● entrepreneurial ● global 70
 Testable Research
Questions(1/2)
• Preceding questions, while unquestionably relevant,
are not testable
• Try to re-cast as testable questions (…even though
the new question may appear less important)
• Scenario…
– You have an idea for a new [technique for entering text on
a mobile phone] and you think it’s pretty good. In fact, you
think it is better than [the commonly used multi-tap
technique]. You decide to undertake a program of empirical
enquiry to evaluate your idea. What are your research
questions?
– Replace […] as appropriate for other research topics

The image part

with www.utm.my innovative ● entrepreneurial ● global 71
 Testable Research
Questions (2/2)
• Very weak (in an empirical sense)
– Is the new technique any good?
• Weak
– Is the new technique better than multi-tap?
• Better
– Is the new technique faster than multi-tap?
• Better still
– Is the new technique faster than multi-tap within one hour of
use?
• Even better
– If error rates are kept under 2%, is the new technique faster
than multi-tap within one hour of use?
The image part
with www.utm.my innovative ● entrepreneurial ● global 72
 Formulating Research
Question
— Research question is an iterative
(repeated) process.
It is start with a set of broad aims or
interest

While literature is searched aims will be

refined, and modified

The Relationship Between the Research

Question, Hypotheses, Specific Aims,
and Long-Term Goals of the Project
The image part 73
with www.utm.my innovative ● entrepreneurial ● global 73
 Steps in formulating
Research Question
• Do I know the field and its literature well?
• What are the important research questions in my field?
• What areas need further exploration?
• Could my study fill a gap? Or lead to greater understanding?
• Has a great deal of research already been conducted in this
topic area?
• Has this study been done before? If so, is there room for
improvement?
• Is the timing right for this question to be answered? Is it a hot
topic, or is it becoming obsolete?
• Will the study have a significant impact on the field?

The image part

with www.utm.my innovative ● entrepreneurial ● global 74
 Research Questions and
Hypothesis
• A process of asking a question, or a
series of related questions, and then
initiating a systematic procedure to
obtain valid answers to that question

The image part

with www.utm.my innovative ● entrepreneurial ● global 75
 Example

The image part

with www.utm.my innovative ● entrepreneurial ● global 76
 Formulating Research
Question/Hypothesis
• A well-thought-out and focused research question
leads directly to your hypotheses.
• Hypotheses are more specific predictions about the
nature and direction of the relationship between two
variables.
• Hypotheses,
– Gives insight into a research question
– Are testable and measurable by the proposed
experiments
• Each hypothesis is matched with a specific aim and
has rationale.
The image part
with www.utm.my innovative ● entrepreneurial ● global 78
 The Thesis Statement
• States your position on a research question
once you’re working on the question
• This is what you “defend” in your defence
• Characteristics
- Reference the research question,
- Unambiguous indication of your view
• “your view” is developed from doing the
research to answer the research question.

The image part

with www.utm.my innovative ● entrepreneurial ● global 79
 Formulating Research
Objectives

The image part

with www.utm.my innovative ● entrepreneurial ● global 80
 Research Objectives
• Objectives are the goals you set out to attain in your study.
– They inform a reader what you want to attain through the
study.
– It is extremely important to word them clearly and specifically.
• Objectives should be listed under two headings:
– a) 1 main objective (aim/purpose);
– b) Several sub-objectives.
• The main objective is an overall statement of the thrust of
your study. It is also a statement of the main associations
and relationships that you seek to discover or establish.
• The sub-objectives are the specific aspects of the topic that
you want to investigate within the main framework of your
study.

The image part

with www.utm.my innovative ● entrepreneurial ● global 81
 Purpose / Aim Statement

ØThe purpose or aim statement shapes the direction of

the research.
ØGives a specific & accurate synopsis of the overall
purpose of the study.

ØExamples of research purpose / aim:

Ø The aim of this study is to examine the existing
risk analysis method and then select the
appropriate solution as the basis for modification
in order to assess and analyze wireless risks
using fuzzy risk analysis method.

The image part

with INSPIRING CREATIVE AND INNOVATIVE
www.utm.my innovative ● MINDS
entrepreneurial ● global 82
 Research Objectives
Ø Covers the different aspects of the problem and its
contributing factors in a coherent way and in a logical
sequence
Ø Clearly phrase the operational terms.
Ø Avoid the use of vague non-active verbs eg to
appreciate, to understand , to study.
Ø Examples of research objectives:
Ø To investigate students’ awareness on information security
and ethical issues within the university.
Ø To evaluate the concept of computer ethics in terms of
information security.
Ø To develop a computer ethics framework focusing on
ethical behavior and information security.

The image part 83

with INSPIRING CREATIVE AND INNOVATIVE
www.utm.my innovative ● MINDS
entrepreneurial ● global 83
 Writing Research Objectives

ØOther example:

Ø To identify the security risks for networked

information systems.
Ø To develop a risk evaluation tool for networked
information systems
Ø To test the evaluation tool for networked
information systems

The image part

with INSPIRING CREATIVE AND INNOVATIVE
www.utm.my innovative ● MINDS
entrepreneurial ● global 84
 Research Objectives
• They should be numerically listed.
• Wording should clearly, completely and specifically
• Communicate to your readers your intention.
– Each objective should contain only one aspect of the Study.
– Use action oriented words or verbs when writing objectives.
• The objectives should start with words such as:
– ‘to determine’,
– ‘to find out’,
– ‘to ascertain’,
– ‘to measure’,
– ‘to explore’ etc.
• The wording of objectives determines the type of research
(descriptive, correlation and experimental) and the type of
research design you need to adopt to achieve them.
The image part
with www.utm.my innovative ● entrepreneurial ● global 85
 Objectives and Research
Questions
• Eg.: To identify…
Objectives

• Answers must be in the related part inside the mind-map of your LR

• The aims and objectives should lead directly to your research
Change to questions.

Research • Change to “question form”

questions

The image part

with www.utm.my innovative ● entrepreneurial ● global 86
 How many research
objectives?

3 objectives
Objective 1 6 months

Objective 2 6 months

Objective 3 6 months

The image part

with www.utm.my innovative ● entrepreneurial ● global 87
The image part
with www.utm.my innovative ● entrepreneurial ● global 88
The image part
with www.utm.my innovative ● entrepreneurial ● global 89
 Example 1: Research
Objectives
i) To propose a new cooperative feature selection scheme that
produces the optimum number and significant of input factors.
ii) To develop hybrid nonlinear-linear model with new sequence of
hybridization that is able to handle incomplete and various size
multivariate time series accurately and robustly.
iii) To investigate the effect of cooperative feature selection,
integration of PSO based BP neural network and changing
hybrid sequence on time series forecasting performance
iv) To evaluate and validate the performance of the proposed
hybrid model with benchmark models on four different set of
time series data that consists of different behavior and scale.

The image part

with www.utm.my innovative ● entrepreneurial ● global 90
 Example 2: Research
Objectives
i) To improve the effectiveness of pre-detection stage by
designing procedures for minimizing unnecessary recognition
and selectively choose the network connection.
ii) To design and develop IDS model that can adaptively learn the
dynamic circumstances in the network traffic and regularly
update the reference model to reflect the changes
iii) To improve discriminative capability of the IDS model to deal
with vague boundary between normal and abnormal traffic
pattern and on imbalanced dataset. In particular this research
investigates on the design of classifiers by the means of
ensemble classifier.

The image part

with www.utm.my innovative ● entrepreneurial ● global 91
 Example 3: Research
Objectives
i) To objectively measure the vulnerability of existing MANET
routing protocol to existing and future types of attacks.
ii) To design, improve and analyze a new lightweight layered
security framework to counter simple attack that is represented
by blackhole attack.
iii) To design, improve and analyze the enhancement of the
security framework to counter sophisticated attack that is
represented by wormhole attack.
iv) To propose and analyze the performance of the security
framework based on regression and neural network data
analysis

The image part

with www.utm.my innovative ● entrepreneurial ● global 92
 Defining Scope

The image part

with www.utm.my innovative ● entrepreneurial ● global 93
 Scope of the Study (1/2)

• Is the boundary of the study

• Domain of the data The imag
relationsh

– Time frame of the data found in th

– Type of data considered

– Attributes to be used
– Area of study
• Techniques to be used

The image part

with www.utm.my innovative ● entrepreneurial ● global 94
 Scope of the Study (2/2)
• Methodology
– Approach
– Testing/experimental tools/measurements
– Test Data
• Specify the limits of the research in a way which
makes it clear what is and is not to be studied,
through for example
– Definitions
– Time spans
– Geographical boundaries Th
pa
– Other limits as appropriate to the field of study re
ID
no
th
The image part
with www.utm.my innovative ● entrepreneurial ● global 95
 Problem Scenario

The image part

with www.utm.my innovative ● entrepreneurial ● global 113
 Example of RQs
RQ1: How can we exploit the concept drift detection methods by leveraging
dissimilarity measures for concept drift detection on user textual reviews to
increase detection accuracy and reduce detection false alarms?

RQ2: How can we incorporate the concept drift of user preference and item
features into a collaborative recommender system based on the temporal matrix
factorization method?

RQ3: How can the accuracy of the recommender system be improved by

utilizing the user sentimental bias filtering and dynamics changes of user and
item features using the deep learning method?
The image part
with www.utm.my innovative ● entrepreneurial ● global 114
 Example of RO
The main goal of this study is to develop models that take into account the concept drift of user
preferences and dynamic item properties for building a robust and reliable RS with high accuracy of
recommendations. Thus, to achieve the goal of the study, the following objectives have been set.

i. To develop a concept drift detection method by leveraging dissimilarity measures for concept drift
detection on user textual reviews in an online sentiment classification task?
ii. To propose a dynamic-based collaborative method by incorporating dynamic changes from user
and item latent features into a temporal matrix factorization to improve the accuracy of the
recommender system.
iii. To propose an adaptive deep recommendation model by utilizing the user sentimental bias filtering
and dynamics changes of user and item features for better predictive performance.

The image part

with www.utm.my innovative ● entrepreneurial ● global 115
 Example of Scope
1. The models were evaluated based on different datasets: Amazon products reviews and
20-Newsgroup datasets were used for concept drift detection method evaluation.
Similarly, Epinions, Baby, and Cell-phone & Accessories datasets were used to
evaluate the rating prediction performance.

2. Regarding the distance measures for drift detection methods, four different measures
were considered: Hellinger distance (HD), Kullback–Leibler divergence (KD), Total
variation distance (TVD), and the Kolmogorov– Smirnov statistic (KS distance).

3. For rating prediction, the Matrix Factorization (MF) and Multilayer Perceptron (MLP)
methods were used for the proposed methods in Chapter 5 and Chapter 6, respectively.

4. To evaluate the performance of drift detection models, False alarm, Miss rate, and
accuracy were used. While the MAE and RMSE metrics will be used to evaluate the
performance of the proposed RS models.
The image part
with www.utm.my innovative ● entrepreneurial ● global 116
 At the end of problem formulation the
researcher should have:
Good literature, and
giving indication of the
theoretical or practical
significance of the study.

Enough Develop testable

background hypotheses that
information in suggest require
the field of further study and
study. experience.
End of
Problem
Formulation
The image part 117
with www.utm.my innovative ● entrepreneurial ● global 117
 Take 10 minutes to write
your objectives
• To develop methods based on CNN for cost-effective feature selection
to be used for accurate classification of imbalanced datasets
• to design methods based on …. To enhance accuracy of electrical
power … .forecasting
• to develop a model that will retrieve the big data from the telecom
platform and present it clearly with 100% accuracy within 8 months
• This study intends to produce an improved and fine-tuned big data
analytics approach, specifically for KPI and dashboarding to help
management in strategic decision making.
• This study also intends to find out the impact of applying big data
analytics during KPI settings over the traditional methods.
• We will also analyse the challenge and pain points in applying big data
analytics for strategic decision making..
• .
The image part
with www.utm.my innovative ● entrepreneurial ● global 118
 THANK YOU

The image part

with www.utm.my innovative ● entrepreneurial ● global 119