Transcript abstractIntroductionUPM

ITENARARY
• 9.00 – 10.00 : Abstract & Introduction
• 10.00 – 10.15 : Rehat
• 10.15 – 11.15 : Related Works & Methodology
• 11.15 -11.30 : Rehat
• 11.30 – 12.30 : Results & Conclusion
• 12.30 – 1.00 : Q & A
• Bersurai
2
OBJEKTIF
• Di akhir Bengkel, peserta akan dapat
• Memahami struktur penulisan makalah yang standard
• Memahami komponen penting dalam setiap bahagian
• Mengaplikasi dalam penulisan makalah sebenar
3
MOTIVATION
PUBLISH THE OUTCOMES
• career progression – moving up to the next rung on the ladder
• gaining recognition for work you have done
• setting yourself a new challenge
• helping your students to gain recognition for their work
• learning how to write to a higher standard
• contributing to knowledge
• Building your institution’s status
• developing a profile
4
CIRI-CIRI PENERBITAN BERMUTU
(Journal)
Journal Selection Criteria (Information Science
Institute, ISI).
The evaluation process consists of evaluation of many criteria
such as, Basic Journal Publishing Standards (including
Timeliness of publication, adherence to International Editorial
Conventions, English Language Bibliographic Information
(including English article titles, keywords, author abstracts,
and cited references.) ISI also examines the journal's
Editorial Content, the International Diversity of it authors and
editors. Citation Analysis using ISI data is applied to
determine the journal's citation history and/or the citation
history of its authors and editors.
5
Susunan abstract/index
Outstanding Index/Abstract
(Engineering, Science & Technology)
•
•
•
•
•
•
•
•
•
•
Science Citation Index (SCI, SCI Expanded)
CompuMaths Citation Index
Current Contents (Engineering, Computing, Technology)
Cambridge Scientific Abstracts
Computer Abstracts UK, CompuSci
Mathematics Review USA
Mathematics Abstracts Germany
DBLP Bibliography
INSPEC
Scopus
503 journals indexed by SCI or SCI Expanded
6
Top 10 Impact Factor (IF) evaluated
by ISI 2007
1
Int. Journal Computer Vision
0920-5691
6.085
2
ACM Trans. Information Sys
1046-8188
5.059
3
Bioinformatics
1367-4803
4.894
4
MIS Quarterly
0276-7783
4.731
5
IEEE Trans. Pattern Analysis
0162-8828
4.306
6
ACM Computing Survey
0360-0300
4.130
7
ACM Trans. Graphics
0730-0301
4.081
8
VLDB Journal
9
Journal Am. Med. Inform. Assn.
1067-5027
3.979
10
IEEE Trans. Evolut. Computing
1089-778X
3.770
N=
IEEE Trans. Know. & Data Eng.
4.062
2.18?
7
OTHERS, IF 2006
1
2
3
Information Systems
4
Data and Knowledge Engineering
5
6
Future Generation Computer
Systems
0.722
8
Information Processing Letters
0.5?
9
ACM PODS.
0.53
10
IJCM, Taylor&Francis
0.428
n
IEICE, Information Systems
0.28
7
8
ISSUES
• no research
• less/no publication – less/no writing
• manuscript below standard
• don’t know how to write
• sub-standard substance
9
The Basics of Good Writing
So you're going to sit down at your computer, go through your notes,
and in a few hours produce a piece of research writing. Right?
Wrong!
It is impossible to start from nothing and produce a good piece of writing,
because it is very hard to organize your material and write at the same
time. If you are working out which piece of research to talk about next
and worrying about verb agreement, you are less likely to produce a good
piece of writing
10
The Basics of Good Writing
Here are some suggestions:
Plan your writing. Before you start writing, find a way to organize your material so
that you know what you are going to write about, in what order, and what you're
going to say. Try writing an outline. Trying writing your ideas down on the back of an
envelope, or a piece of old paper. It doesn't have to be beautiful, it just has to help
you think about what you are going to say. Use whatever method works for you, no
matter how strange!
Ignore the language! When you plan your writing, don't worry about the language.
Concentrate on what you are going to say. Write in notes so that you don't have to
think about verb agreement. Don't waste time worrying about spelling. You can think
about all these aspects of writing after you've decided what you are going to say. If
you spend a lot of time fixing all the prepositions and conjunctions in an early draft
you are not going to be willing to cut out paragraphs or sentences that you later
realize aren't necessary, or to change them substantially. So don't put a lot of effort
into proofreading until you are sure that what you want to say is the best you can
come up with, then you can spend more time on fixing up the writing so that the
punctuation, spelling, etc. is correct.
11
The Basics of Good Writing
Write and rewrite! More experienced writers rewrite more times and more
substantially than less experienced writers. Are you surprised? Good writing takes time
for everyone. The better a writer you become, the more you will see that the first
thoughts/ideas/writing that comes out of your head and onto the page can be
improved. So give yourself time to rewrite so that your readers see the best of your
thoughts and writing, not the best you could come up with at the last moment.
Find readers! ask people to read what you've written. Ask friends, ask professors, ask
your writing advisor in languages. But don't wait until your writing is "perfect" because
then if people suggest changes you won't want to make them! Give people drafts and
let them know what sort of feedback you want: comments on organization? on ideas?
on your language? on the technical aspects of what you've written?
Keep writing! Good writing takes practice. The only person who can make you a
better writer is you. So work at it, show your work to other people, and rewrite,
rewrite, rewrite.
12
WRITING FORMAT
A COMMON FORMAT IN WRITING COMPRISES OF:
•
•
•
•
•
•
•
•
•
Abstract
Introduction
Related Work / Literature Review
Material & Methodology
Results
Discussion
Conclusion
Acknowledgement
References
13
ABSTRACT
An abstract should briefly:
•
•
•
•
•
Re-establish the topic of the research.
Give the research problem and/or main objective of
the research (this usually comes first).
Indicate the methodology used.
Present the main findings.
Present the main conclusions
14
ABSTRACT
The Body of the Abstract
The abstract is a very brief overview of your ENTIRE
study. It tells the reader WHAT you did, WHY you
did it, HOW you did it, WHAT you found, and WHAT
it means.
Briefly state the purpose of the research (introduction),
how the problem was studied/solved (methods), the
principal findings (results), and what the findings mean
(discussion and conclusion). It is important to be
descriptive but concise--say only what is essential,
using no more words than necessary to convey meaning.
15
Common Problems
Too long. If your abstract is too long, it may be rejected - abstracts are
entered on databases, and those is usually a specified maximum number
of words.
Too much detail. Abstracts that are too long often have unnecessary
details. The abstract is not the place for detailed explanations of
methodology or for details about the context of your research problem.
Too short. Shorter is not necessarily better. If your word limit is 200 but
you only write 95 words, you probably have not written in sufficient
detail. Many writers do not give sufficient information about their findings.
Failure to include important information. You need to be careful to
cover the points listed above. Often people do not cover all of them
because they spend too long explaining, for example, the methodology
and then do not have enough space to present their conclusion.
16
TITLE
The title should convey to the objective/purpose of the paper. The
nicer the better.
Example:
A Support-Ordered Trie for Fast Frequent Itemset Discovery
17
ABSTRACT
Model 1
The importance of data mining is apparent with the advent of
powerful data collection and storage tools; raw data is so abundant that
manual analysis is no longer possible.
Unfortunately, data mining problems are difficult to solve and this
prompted the introduction of several novel data structures to improve
mining efficiency.
Here, we will, critically examine existing preprocessing data structures
used in association rule mining for enhancing performance in an attempt
to understand their strength and weaknesses.
Our analysis culminate in a practical structure called the SOTrieT
(Support-Ordered Trie Itemset) and two synergistic algorithms to
accompany it for the fast discovery of frequent itemsets.
Experiments involving a wide range of synthetic data sets reveal that its
algorithms outperform FP-growth, a recent association rule mining
algorithm with excellent performance, by up to two orders of magnitude
and, thus, verifying its efficiency and viability.
18
ABSTRACT
Model 2
The importance of data mining is apparent with the advent of
powerful data collection and storage tools [WHY].
Unfortunately, data mining problems are difficult to discover the
frequent
itemsets
due
to
computationally-intensive
process
[WHY/WHAT].
This paper will propose a model called SOTrieT (Support-Ordered Trie
Itemset) and two synergistic algorithms accompany it for the fast
discovery of frequent itemsets [HOW].
Experiments involving a wide range of synthetic data sets reveal that its
algorithms outperform FP-growth, a recent association rule mining
algorithm with excellent performance, by up to two orders of magnitude
and, thus, verifying its efficiency and viability.[WHAT YOU FOUND]
19
INTRODUCTION
The introduction comes at the start of a piece of
writing. It introduces the research by situating it (by
giving background), presenting the research problem
and saying how and why this problem will be "solved."
Without this important information the reader cannot
easily understand the more detailed information about
the research that comes later in the paper. It also
explains why the research is being done (rationale)
which is crucial for the reader to understand the
significance of the study.
20
PURPOSE OF THE INTRODUCTION?
What is the context of this problem? In what situation or environment can
this problem be observed? (Background)
Why is this research important? Who will benefit? Why do we need to
know this? Why does this situation, method, model or piece of equipment need
to be improved? (Rationale/justification)
What is it we don’t know? What is the gap in our knowledge this research
will fill? What needs to be improved? (Problem Statement)
What steps will the researcher take to try and fill this gap or improve the
situation? (Objectives)
Is there any aspect of the problem the researcher will not discuss? Is the study
limited to a specific geographical area or to only certain aspects of the
situation? (Scope)
21
INTRODUCTION
COMMON PROBLEMS
Too much detail, and hence too long. Remember, this is the introduction, a
kind of overview. Although you will cover important points, detailed descriptions of
method, study site and results will be in later sections. Look at the proportion of a
research paper an introduction takes up. Notice it is comparatively short because it
serves as a summary of what follows.
Repetition of words, phrases or ideas. You will have keywords that are crucial
to your study. However, your reader doesn't want to read them over and over! A
high level of repetition makes your writing look careless. To reduce it, highlight
repeated words or phrases - then you can easily judge if you are overusing them
and find synonyms or pronouns to replace them.
22
COMMON PROBLEMS
Unclear problem definition. Without a clear definition of your research problem,
your reader is left with no clear idea of what you were studying. This means that
they cannot judge your work's relevance to their own work, or its usefulness, quality,
etc. As an exercise, you should be able to complete a sentence that starts, "The
purpose of this study is . . . " that encapsulates the problem you are
investigating. Of course you will not include this exact sentence in your paper, but it
serves as an easy way to check that you have a clearly defined problem. In your
paper you should be able to write your research problem in one sentence - you can
add details in the sentences that follow. You should also ensure that your research
problem matches the title of your paper (you'd be surprised how many don't !) as
well as its methodology and objectives.
Poor organization. Writing an introduction that effectively introduces your research
problem and encapsulates your study is not an easy task. Often when we write we
discover gradually what we want to say and how we want to say it. Writing is often a
process of discovery. Bear this in mind when you write your introduction, and be
prepared to go back and make big changes to what you have written, and the order
in which you have presented your ideas and information. Your introduction must have
a logical sequence that your reader can follow easily.
23
A SCHEMA FOR INTRODUCTIONS
• Establish the Field – centrality, general to specific, previous research
• Define a research problem – gap, questions, previous inquiry
• Propose a Solution
24
FIELD
First you need to establish the area of research in which
your work belongs, and to provide a context for the
research problem. This has three main elements:
Claiming centrality: Claiming that the area of research is an
important one, and therefore implying that the research done is also
crucial.
For example:
“ … although computing power has increased tremendously over the
years, efficient algorithms with customized data structures are still
necessary to obtain timely results. “B. Cui et. al. July 2004, IEEE
Trans. On Knowledge and Data Engineering” Here the words "
efficient algorithms with customized data structures " indicates
centrality by showing that this factor is crucial.
25
FIELD
General to specific: Most writing starts with general
information and then moves to specific information. This
is true of introductions too.
General
Specific
26
FIELD
General to specific: For example:
… This prompted keen interest in automated data analysis
tools, that in turn, catapulted the rise of data mining. However,
data mining process with the current data structure having a
computationally-intensive process i.e. NP-complete.
We focus on Association Rule Mining (ARM) because of its
immense popularity and usefulness in a wide variety of
situations such as e-commerce, classification, clustering, webmining, and bioinformatics.
Several data structures have been proposed for the
enhancement of ARM, but none is able to cope effectively with
the size and dynamism of current database.
27
FIELD
Explaination
The first sentence: This prompted keen interest in automated data analysis tools,
that in turn, catapulted the rise of data mining. (general).
The second sentence: However, data mining process with the current data
structure having a computationally-intensive process i.e. NP-complete. (less
general).
The third sentence: We focus on Association Rule Mining (ARM) because of its
immense popularity and usefulness in a wide variety of situations (more specific).
The fourth sentence: Several data structures have been proposed for the
enhancement of ARM, but none is able to cope effectively with the size and
dynamism of current database. (yet more specific).
28
FIELD
Previous research: Often the introduction will refer to work already done in the
research area in order to provide background (and often also to help define the
research problem). For example:
Several data structures have been proposed for the enhancement of the
ARM, but none is able to cope effectively with the size and dynamism of
current databases
[9-14].
29
DEFINE RESEARCH PROBLEM
PROBLEM
Your research must be new in some way. It must add
knowledge to your field so you need to show in what way your
work explores an area/issue/question that has previously not
been explored, or not been explored in detail, in not explored
in the way that you are going to use. In other words, you
need to give a rationale for your work (i.e. show the reasons
for doing it). There are four ways to demonstrate that you are
adding to the knowledge in your field:
30
DEFINE RESEARCH PROBLEM
Gap: A research gap is an area where no or little research has been
carried out. This is shown by outlining the work already done to show
where there is a gap in the research (which you will then fill with your
research).
For example:
The structures that had been proposed for mining the data were
not able to cope effectively with the size and dynamism of
current databases.
31
DEFINE RESEARCH PROBLEM
Raising a question: The research problem is defined by asking a
question to which the answer is unknown, and which you will
explore in your research. For example:
The question we address here is how technological change occurs
when it is the overall system that needs to be changed. In
particular, how can we begin and sustain a technological transition
away from hydrocarbon based technologies? (Street and Miles,
1996)
32
DEFINE RESEARCH PROBLEM
Continuing a previously developed line of enquiry: Building on
work already done, but taking it further (by using a new sample,
extending the area studied, taking more factors into consideration,
taking fewer factors into consideration, etc.). For example:
Taking all these elements and their possible variations into account is
often far too complex and tedious for determining efficient gas
development patterns with simple back of the envelope calculations. In
their survey of these elements, Julius and Mashayeki [8] present a
detailed analysis of these different interactions. They suggest that these
be taken into account through gas planning models constructed in the
same spirit as the planning models developed in the power generation
sector.
In this paper, we present a gas planning model that fulfils some of the
specifications established in Julius and Mashayeki [8]. (Boucher and
Smeers, 1996)
33
DEFINE RESEARCH PROBLEM
Counter-claiming: A conflicting claim, theory or method is put
forward. Here, for example, the researchers argue that previous
researchers' assessments of cost effectiveness/methodology were too
complex, and that a simplified process could and should be used instead:
34
PROPOSE A SOLUTION
SOLUTION
Once the field and problem have been defined, it is time to give the
"solution." In other words, how will the research gap be filled? How
will the question that was raised be answered?
This last part of the introduction can also be used to show the
benefits, to explain the objectives, to clarify the scope of the research,
to announce what was found from doing the research and how it can
be used. Notice that an introduction will discuss a number of the
following points but is unlikely to cover them all.
35
PROPOSE A SOLUTION
Outlining purpose: Often researchers will described their objectives in
their introduction in order for the reader to have a clear idea of what they
set out to accomplish. Usually there is a general objective written in one
sentence (details of more specific objectives can be given in following
sentences). For example:
This work aims to establish the extent of interaction of alginate with
calcium and aluminium ions with respect to the influence of algal exudates
have on the removal of humic substances by aluminium coagulation during
drinking water treatment. (Gregor et al., 1996)
Hint!: always give an overall objective before giving specific
objectives. This will help you explain much more clearly to your
reader what your work aimed to accomplish.
36
PROPOSE A SOLUTION
Announcing present research (method): Important points about
the methodology used are outlined, perhaps including the scope of
the study. However, the methodology is not given in detail (since
details are given in the methodology section). For example:
In this paper, we focus on ARM…..Here, we shall critically examine
existing preprocessing data structures in ARM for enhancing
performance in an attempt to understand the strengths and
weaknesses. Our analysis culminate in a practical structure called
Support-Ordered Trie Itemset (SOTrieIT) and two synergetic AR
mining algorithms to accompany it. Experiments involving a wide
range of synthetic data sets reveal that the algorithm outperform
FP-growth, a recent ARM algorithm by up to two orders of
magnitude…..
“B. Cui et. al. IEEE Trans. On Knowledge and Data Engineering, pp 875879, July 2004. ”
37
PROPOSE A SOLUTION
Example:
This paper proposes a new model called Support-Ordered Trie
Itemset (SOTrieIT) with synergistic ARM algorithms to improve the
performance process.
38
PROPOSE A SOLUTION
Announcing principle findings (results): Researchers may
indicate the kind of results they obtained, or an overall summary of
their findings. For example:
Different operating modes of the MESFET mixers, gate mixers, drain
mixers, and resistive mixers were investigated in this work and the
results proved that good conversion characteristics could be
achieved.(Angelov, 1991)
39
PROPOSE A SOLUTION
Example:
It reveals that the performance of the model increases up to two
orders of magnitude and, thus, verifying its efficiency and viability.
40
PROPOSE A SOLUTION
Indicating the structure of the research: It is useful to
outline the organization of the written up research that follows
so that the reader has a clear idea of what is going to follow,
and in what order. In other words, “Paper Organization”
For example:
The rest of the paper is organized as follows: The next section
describes the problem of mining ARs. Section 3 reviews the
ARM algorithms that utilize data structures, while section 4
introduces our enhanced data structures and algorithms. The
performance of our approach is evaluated in section 5, and
finally, the papers is concluded in section 6.
“B. Cui et. al. IEEE Trans. On Knowledge and Data Engineering, pp 875879, July 2004.”
41
PROPOSE A SOLUTION
Indicating directions for further research: Research often
opens up other areas where research could or should be done,
so it is common for these areas to be defined in the
introduction. It is also a way of indicating that the current
study is not designed to be comprehensive.
This paper takes a first step in this direction by laying out the
rationale for incorporating feedback and feedforward
mechanisms in decision support for dynamic tasks such as
software project management (Sengupta and Abdel-Hamid,
1993).
42
PROPOSE A SOLUTION
Indicating benefits of current research: Indicating the
benefits of the research helps to justify why it was carried out
and emphasizes the value of the study. For example:
…..because of its immense popularity and usefulness in a wide
variety of situations such as e-commence, classification,
clustering, web mining, and bio-informatics.
43
PROPOSE A SOLUTION
Notice that the introduction includes information that is
presented in other parts of the paper. Does this mean
that if you indicate your results in your
Introduction that you will have nothing left to present
in your Results chapter?
No! Introductions literally "introduce" information to
give an overview, often offering only a short summary
because full details are given in later chapters.
44