Research TOPIC & Research

PROBLEMS
Prof Dr Adam Amril Jaharadak
Learning outcomes

To generate ideas that will help in the choice of a suitable research

topic

To define a proper research problem

MSU Key research area (Technology)

NEW MEDIA CYBERSECURITY BIG DATA ENGINEERING SUSTAINABLE ENERGY

TECHNOLOGY INNOVATIVE
SOLUTIONS
What is Research?
• Research is all about solving problem depending on environment or
solving unanswered question
• It is about finding out about something. For example, in
biotechnology, we have genetic engineering and many more
• When you do a research actually you want to answer a question
• Research can be described as interest to pursue to find out something
which is useful and can contribute whether in biological, physics or
any field of interest.
What is Research?

Research is when you want to

seek knowledge.
Solution to problems.
Way of advancing knowledge.
Typical Research Process for students
Research Topic
• Types of Research Topic
• • Unanswered question • Unsolved question • Concern • Query
• • Statement of • Personal interest • Social problem • Testing theory •
Prior research • Program evaluation • Human service practice •
Minorities in research inquiry
Criteria for Selecting a Topic
• • Relevance: The topic you choose should be a priority problem:
Questions to be asked include:
• • How large or widespread is the problem?
• • Who is affected?
• • How severe is the problem?
• • Avoidance of duplication: Investigate whether the topic has been
already researched
• • Feasibility: Consider the complexity of the problem and the
resources you will require to carry out the study.
Criteria for Selecting a Topic
• • Political acceptability: It is advisable to research a topic with the
authorities' interest and support.
• • Applicability of possible results and recommendations
• • Urgency of data needed
• • Ethical acceptability – religious issues, racial conflicts, cast problems
How to choose your topic?

• Something you are capable of undertaking

• Something that excites your imagination
• Something that you have a genuine interest in
• Something that develops research skills as part of undertaking
your project
• Something in which you will not lose interest while developing
the project
• Something you are capable of receiving data for
Techniques for generating research ideas

Rational thinking: Creative thinking:

• Examining your strengths and •Keeping a notebook of ideas
interests •Exploring personal
• Looking at past project titles preferences using past
projects
• Discussion
•Relevance trees (mind
• Searching the literature mapping)
(articles, reports, books) •Brainstorming
Approach for research ideas

• One way to define your topic is to select a broad topic, then identify
one or more sub-topics you might like to explore.
• Broad topic: e.g. Artificial Intelligence
• Sub topic: e.g. Supervised vs. unsupervised machine learning
• Sub topic: e.g. Model-free & model-based reinforcement learning
algorithms
• Sub topic: e.g. Computer vision application: autonomous cars
Approach for research ideas

• Another approach is to select a topic, then list possible questions, such as...
• E.g. Topic : Artificial Intelligence
• Who? E.g. Who will benefit the most from reinforcement learning?
• What? E.g. What areas of life machine learning are the most influential?
• Where? E.g. Where do recommender systems’ approaches optimally
functioned?
• When? E.g. When to choose between natural language processing and
machine learning?
• Why? E.g. Why do the Internet of things devices generate so much data?
• How? E.g. How do social robots interact with humans?
Using library research
• Library research involves the step-by-step process used to gather
information in order to write a paper, create a presentation, or complete a
project.
• As you progress from one step to the next, it is commonly necessary to
back up, revise, add additional material or even change your topic
completely.
• This will depend on what you discover during your research.
• There are many reasons for adjusting your plan. For example, you may find
that your topic is too broad and needs to be narrowed, sufficient
information resources may not be available, what you learn may not
support your thesis, or the size of the project does not fit the requirements
Sources of ideas
• Primary sources are original works. These sources represent original
thinking, report on discoveries, or share new information. Usually
these represent the first formal appearance of original research.
Primary sources include statistical data, manuscripts, surveys,
speeches, biographies/autobiographies, diaries, oral histories,
interviews, works or art and literature, research reports, government
documents, computer programs, original documents(birth
certificates, trial transcripts...) etc.
Sources of ideas
• Secondary sources are studies by other researchers. They describe,
analyze, and/or evaluate information found in primary sources. By
repackaging information, secondary sources make information more
accessible. A few examples of secondary sources are books, journal
and magazine articles, encyclopedias, dictionaries, handbooks,
periodical indexes, and reviews, etc
Activities 1: Brainstorming (10-15 minutes)

Given a problem - Ask questions relating to the problem

NO SUGGESTIONS SHOULD BE CRITICISED OR EVALUATED before all ideas have

Record all suggestions, observing the been considered-

following rules All suggestions should be recorded and considered-

As many suggestions as possible should be recorded

Review all suggestions and explore what is meant by each

Analyze the list of suggestions and decide which appeal to you as research ideas
and why
Problem:

• Low accuracy of detection in current brain

tumor prediction system
Developing research title
• A research title is a product of real world observations, dilemmas,
wide reading, selective viewing, films documentaries and deep
reflection. The title of the research is the research problem in capsule
form. It must clearly reflect the topic of investigation and it must be
original, clear, concise or specific
Developing research title
• Writing the Title
• ■ The title gives a name to the research and it is also a means to
identify the contribution of the researcher
• ■ It should be clear enough to express the global content of the
article in a brief and concrete way
Developing research title
• Considering the Reader
• ■ The title is the first thing that the readers meet.
• Therefore, it should be attractive and create a good impressions

• Length and Key Words

• ■ The title should be no longer than 15 words and should summarize
the main topic of the report, and identify the crucial issues (key
words).
Developing research title
• The following are some guidelines in writing the research title
• The research title sums up the variable being studied in the
research.
• Usually, the general problem is reflected in the research title
• Avoid titles that are too long.
• A good research title is usually composed of 12 to 15 words
 Example of broad Topics

3. Cloud
1. Smart City 2. Heart Disease 4. Electronic
Computing
Cyber Security Prediction System Health Records
Services

6. Student 7. Big Data in

5. IoT for River 8. IoT for City
Performance Hospitality
Sustainability Surveillance
Prediction System Industry
Developing research title
• Example of Specific Topics
• Remote monitoring information system and its applications based on
the Internet of Things.
Activity 2: Using any topic given, propose a good
Proposing research title using the guide that you have
learned
research title
What is a research problem?
Ø A research problem - refers to some difficulty which a researcher
experiences in the context of either a theoretical or practical
situation and wants to obtain a solution for the same
Ø Like a medical doctor, a researcher must examine all the symptoms
(presented to him or observed by him) concerning a problem before
he can diagnose correctly.
Ø To define a problem correctly, a researcher must know: what a
problem is?
Problem Statement
Ø Should include
Øa clear statement that the problem exists,
Øevidence that supports the existence of the problem,
Øevidence of an existing trend that has led to the problem,
Ødefinitions of major concepts and terms (this can be provided below in a
subsection),
Øa clear description of the setting,
Øprobable causes related to the problem, and
Øa specific and feasible statement.
 Flow of Ideas in the Problem Statement
Subject area

• Concern or issue
• How will addressing what • A problem
we need to know help • Something that needs
researchers, educators, a solution
policy makers, and other
individuals?

• Evidence from the

literature
• In this body of • Evidence from
evidence what is practical experience
missing or what do we
need to
know more about
 Examples Of The Problem Statement
Ethical issues
in colleges

Ethical violations among

• Assessing violations
football recruiters
•Helps recruiters develop
better ethical standards
• Helps athletes understand
ethical issues

•Gap in the literature

Description identifying •Reports of violations
and characterizing
violations
Why is the Research Problem Important?
Ø It establishes the importance of the topic.
Ø It creates reader interest.
Ø It focuses the reader’s attention on how the study will add to the
literature.
How the Problem Differs From Other Parts of
Research?
• • A research problem is an educational issue or problem in the study.
• • A research topic is the broad subject matter being addressed in a
study.
• •A purpose is the major intent or objective of the study.
• • Research questions are those that the researcher would like
answered or addressed in the study.
Differences among the Topic, Problem, Purpose,
and Questions
• Distance learning
General Topic

• Lack of students in distance classes

Research
problem

• To study why students do not attend distance education

Purpose classes at a community college
Statement

Specific • Does the use of website technology in the classroom

deter students from enrolling in a distance education
Research
Question class?
Example: Topic & Title

• General Topic : Image Forensic

• Specific Title : Image Splicing Detection using Mean Shift and
Normalized Cuts Algorithm
Example: Background study
• Living in current age, people are exposed to a remarkable array of
visual imagery.
• Doctored photographs are appearing with a growing frequency
and sophistication.
• Fast development of commercial image editing software e.g. Adobe
Photoshop dramatically increases the amount of doctored
photographs circulated every day.
• Leads to serious consequences, reducing trustworthiness and
creating false beliefs in many real
• Composition of a great white shark leaping out of water attacking military
personnel climbing to a helicopter (Danielson, 2020)
Example: Problem statement
• Bottleneck cause by automatic segmentation technique
• Tampering detection systems should be fast and efficient, meeting
the realistic constraints on computational speed and memory usage.
In the current consistency checking technique, one speed bottleneck
is caused by automatic image segmentation, which takes a few
minutes to complete for a typical sized image of 800x600 pixels (Hsu
& Chang, 2020).

• Remaining issue:
• Speed bottleneck caused by automatic image segmentation
Example: justification
• Traditionally, image forensics has been done with human inspection
• Achieve accurate detection and high quality analysis
• BUT require significant amount of time and extensive human labour
• Therefore, automated content integrity verification is highly
required
• For fast verification processes
• Complement human inspection for manipulations that cannot be
perceptibly detected by the human eye
Example: Goal Of Study
• To develop a single image tampering detection framework, to solve
image level detection problem using a consistency checking method
based on imaging device characteristics
Example: Research Objectives
• To improve the speed caused by automatic image segmentation
using combination of Mean Shift (MS) and Normalized Cuts (NCuts)
algorithm.
• To propose the key factor that contributes to the success of splicing
detection.
Example: Research Questions
• How can someone confirm that the image is authentic (trustworthy)
or has been doctored (cannot be trusted)?
• What are the key factors that contribute to the success of the
splicing detection?
• How fast the technique detect and confirm that the candidate image
is spliced?
Activity 3: Understanding Problem
• Find a recent journal articles (within 2020-2024) that have the
information content related to your research area of interest. Read
the articles and answer the following questions:
1. What is/are the problem(s) being investigated in those articles?
Briefly describe for each article.
2. How does the problem being overcome? Describe the approach /
method / experiment being used.
3. What are the remaining issues (gap) that is/are still not been
resolved? Highlight concisely.
4. List down all the keywords that describe the above problem(s)
An Intrusion Detection System for SDN Using Machine Learning

Abstract:
Software Defined Networking (SDN) has emerged as a promising and exciting option for the future
growth of the internet. SDN has increased the flexibility and transparency of the managed,
centralised, and controlled network. On the other hand, these advantages create a more vulnerable
environment with substantial risks, culminating in network difficulties, system paralysis, online
banking fraud, and robberies. These issues have a significant detrimental impact on organisations,
enterprises, and even economies. Accuracy, high performance, and real-time systems are necessary
to achieve this goal. Using a SDN to extend intelligent machine learning methodologies in an
Intrusion Detection System (IDS) has stimulated the interest of numerous research investigators
over the last decade. In this paper, a novel HFS-LGBM IDS is proposed for SDN. First, the Hybrid
Feature Selection algorithm consisting of two phases is applied to reduce the data dimension and to
obtain an optimal feature subset. In the first phase, the Correlation-based Feature Selection (CFS)
algorithm is used to obtain the feature subset. The optimal feature set is obtained by applying the
Random Forest Recursive Feature Elimination (RF-RFE) in the second phase. A LightGBM algorithm
is then used to detect and classify different types of attacks. The experimental results based on the
NSL-KDD dataset show that the proposed system produces outstanding accuracy, precision, recall,
and f-measure compared to the existing methods.
 •Q & A

Transforming Life, Enriching Future

Thank you!