The best method for presentation of research results

in theses and papers

Ivan Stojmenovic [email protected]

This text describes how to write research papers or research Ms. or Ph.D. theses. Most of the
instructions are also applicable for preparing transparencies (because of some differences,
transparency preparation is explained in a separate document, written by Veljko Milutinovic,
copied also on my web page in parallel to this document). If this method is not suitable for
your article, chances are that you are not doing research but rather development,
implementation or other type of work.

The main key to the successful presentation is to repeat your ‘story’ four times: in the title,
abstract, introduction (or chapter 1) and in the text. That is, make readable and as complete as
possible versions of your work using the order of 10, 100, 1000 and 10.000 words. Why?
Because you have four different types of readers. Among those who will ever notice your
work, I estimate that 80% will see only title, 15% will read the abstract, 4% will read also
introduction, and the surviving 1% will read the whole paper. This is assuming you did your
job properly, otherwise you will have 0% in the last group. If you do not believe this, just
make your own statistics on things you read, and time you have (or time you want to spend)
for reading other people’s work.

Title
Choose a title that enables the expert to figure out the essence of the basic idea(s) and the
main contribution(s).

Wrong title may bring you wrong referees or examiners, and may not attract busy colleagues
to your work when seeing it on your web page or on a search engine that may not even catch
your work (which decreases your visibility). If you are solving problem X using method Y,
you have probably searched Internet for both X and Y. Others do the same, so let your work
be observed easily.

Abstract
State clearly what problem has been studied and/or what is the goal of the thesis/paper. Give
a brief statement on existing solutions and their drawbacks. List major contributions of the
thesis. State briefly assumptions and limitations. The abstract should also include major
idea(s), the type (e.g. performance, complexity) and result of analysis done.

The abstract is written for researchers that are familiar with the research area, and can grasp
your contribution easily. Most of them have worked on the same or related problem. Clear
abstract is the key to having your work properly credited in other people’s work, since again
colleagues may be too busy to figure it themselves, and are more likely to ignore it then to
spend time doing the work you were supposed to do. Examiners will also have a friendly
start with your text. Misleading abstracts are unfortunately quite common practice in the
research literature. Avoid excessive explanations that should be part of introduction. First
answer above questions, then see whether you have space to say anything else.

This structure is also suitable for performance evaluation type of articles. In a performance
evaluation article, the problem is to determine the best protocol under various conditions.
Existing performance evaluations are existing solutions etc. Survey type of articles have
different presentation style. A survey should describe all relevant solutions, classify them
according to assumptions made and some properties (that is, present a taxonomy), and draw
some conclusions.

Chapter 1 (Introduction)

The introduction of the paper, or Chapter 1 of the thesis, should give the summary of the
article. It should contain separate sections on the following items:

i) Introduction (basic facts needed to tune the reader to the thesis or paper);
ii) Problem statement (precise definition and importance); avoid very technical
definitions and statements (present them in later text) and instead give good intuition
for your involved definitions or facts.
iii) Existing solutions and their criticism (limit only to those directly relevant to the
contribution of the thesis; give a motivation for doing research on the topic);
iv) Contributions (proposed solutions; why they are expected to be better; essence of the
idea(s) used in proposed solutions);
v) Conditions, assumptions and limitations of the research done;
vi) Analysis (theoretical, experimental, simulations, implementations,…) done in thesis;
under what conditions and scenarios is your solution best?

In case iii, iv, vi, give only the highlights, with pointers to later sections and chapters that
provide details. The introduction should attempt therefore to present full version of your
article in readable and intuitively clear form. Many of your reviewers, and, with significant
probability even examiners, will stop reading your article unless you fully convinced them in
introduction (or Chapter 1) about your contribution. Life is demanding, and they have other
tasks to do. Readers and especially followers of your research direction will appreciate such
style and will prefer mentioning your work rather than the work of someone else who
remained unclear in introduction and the article was never read due to the lack of time.

Literature review

Chapter or section 2 should give a full literature review. It should collect all known results
relevant to the problem stated, whether or not they are used in proposed contributions. No
additional literature review shall be added in later chapters. In later chapters, you may only
refer to well known results (e.g. those covered in undergraduate computer science program
such as Dijkstra's shortest path algorithm, sorting algorithms etc.). Discuss advantages and
drawbacks of known solutions that are relevant to your problem, and also discuss the
relevance of each reviewed item to your topic and your solutions.
It is very easy for a reviewer or even examiner to save his time by observing a missing
important reference, and claiming that that reference may solve your problem in a better way.
That may or may not be true, but you can loose even if your solution is better, since decisions
made are in most cases final, and your publication or even defense can be prolonged with or
without good reasons.
For every discussed reference, it is very important to relate them to your problem and
contribution in one of several ways: it does not exactly solve the same problem, it solves the
same problem but makes different assumptions about the system, it has some limitations that
you do not have, it makes the same assumptions but does not work well under certain
conditions and scenarios for which you have better solutions, or, if none of these is true, you
are considering it as valid competitor, and will try to defeat it in your analytical or
experimental comparisons. If you are not able to defeat it (under some assumptions and/or
scenarios), I would advice you not to go public with your research and earn undesirable
reputation.

The remaining chapters (sections)

The remaining chapters should present your contributions (including conditions,

assumptions, and limitations, where appropriate), and their analysis. That is, very same items
listed above shall be presented in full, preferably in the same order.
Many problems have complex formulation or are based on terms that a reader is supposed to
be already familiar with. You need to make sure that the audience is as wide as possible,
given the space you have to express yourself. If the problem formulation is somewhat
complex, make sure you give an illustration about the problem. Draw a figure if appropriate,
clearly state what is the input in that figure, what is the expected output in that case. If you
use some terms specific to an area, try to enlarge audience by defining them. Few extra
sentences might be needed by a reader (examiner) generally working in the area, but not
particularly doing research on similar problems. Make sure your article is readable with
ordinary efforts; otherwise it may not be read at all.
Your proposed solution may also need to be nicely illustrated. First give (or repeat, if already
done in introductory parts) the intuitive solution. Then go into details of your solution, step
by step. A figure is worth a thousands words. Is your solution always working? Can you
prove it? How does it work? Is your analysis about worst case or average case? What kind of
theoretical and analytical support can you give for your proposed scheme/solution?
One of the most frequent mistakes made by authors is to ignore the limitations, assumptions
made (compared to assumptions made in other relevant papers), and scenarios for which their
solution is clear winner. You shall not be overly optimistic about your solution(s) and make
unfounded claims. Smaller but justified claim is better than large unfounded claim. Your
referees may turn easily down your paper because of unsupported large claim, but can also
easily accept your even minor contribution if it is well documented and proven. Include all
the possible criticism of your own contribution directly in your article. It is much better that
you criticize yourself than to leave such ‘pleasure’ to examiners and referees. Show that you
are in full control of the problem and solutions.
The performance of your solution can be compared with existing solutions, if any exists
under same or similar assumptions, analytically and/or by doing a simulation. In case of a
simulation/experiments, make sure you give sufficient information to the reader so that
anyone wishing to repeat the experiments will be able to do so and obtain similar
experimental results. Make sure you selected the best independent variables and measured
the performance by key indicators. Select key diagrams or tables to present. Too many
diagrams may show that you are lost yourself in your simulation rather than showing really
important information. Many less important tables/diagrams can be replaced by few
sentences giving the key derived information from the experiments. Give some explanation
for the obtained results. The reader needs some assurance that your program did not have
bugs. The best way is to explain why such kind of data are obtained.

Conclusion

What did you achieve with this research? What are the drawbacks of your solution(s)? What
kind of future work can be done? Do you have some ideas that you intend to study further?
The ownership of some other possible solutions, not fully explored, or subject of your
forthcoming different article, can be protected by outlining them briefly in the conclusion
section, sometimes with reference to upcoming article.
To conclude this advise, try to follow a +-+ pattern in introduction and main text. That is,
start with positive enthusiastic comments about your work and contribution, then become
realistic and list all the drawbacks and limitations, but then finish on a positive note, with a
clear winner statement about the value of your contribution. It is important that the reader
stops reading your article with positive impression. He might be writing his report
afterwards.
Finally, it is very important to use proper English grammar and sentence structure. Ask for
help if your English is not up to the required standard. You must be very careful with
misprints. Do read your article carefully one more time, after some time delay, and check for
possible misprints. The referees and examiners expect you to be very professional. They are
not robots, they are human, and their opinion is partially subjective. Try to make positive
value for the subjective part in overall evaluation by showing that you take care of your
writing.

Any comments, criticism, and additions to this advice on writing articles are most welcome.
Please send them to [email protected].