How to Read &Write Good Technical Papers?

Presented By

Dr. Mohammad Mehedi HassanAssistant Professor

Information Systems Department, CCISKing Saud University, Riyadh, KSA

Email: mmhassan@ksu.edu.saPhone: 4695202

Outline• What is a technical paper?• How to read a technical paper?• Factors affecting writing a Good Paper?• Contents of a Technical Paper• Writing Style• Strategies in Writing a “Good” Paper• Summary

2

What is a technical paper?

• A brief and to-the-point document which describes a body of technical work

• There are many types of technical papers– Survey papers– Journal papers– Conference papers– Letters

3

How to read technical papers?• Should not read everything• Know what’s going on (e.g., scanning

proceedings)– title, authors, abstract

• Papers in your broad research area– introduction, motivation, solution description,

summary, conclusions– sometimes reading more details useful, but not

always

• Papers you may want to improve on– read entire paper carefully

4

What to note after reading?

• Authors and research group– Need to know where to look for a paper on

particular topic

• Theme of the solution– Should be able to go back to the paper if you

need more info

• Approach to performance evaluation• Note any shortcomings

5

Papers Communicate Ideas

• Our goal: to infect the mind of our reader with our idea, like a virus

• Papers are far more durable than programs

6

Fallacy: The greatest ideas are (literally) worthless if you keep them to yourselfFallacy: The greatest ideas are (literally) worthless if you keep them to yourself

Factors affecting writing a Good Paper?

• Good Research Results• Time• Need to be ‘focus’• Careful Planning and Need to be‘Creative’• Organization of the Paper• Writing Styles• Language

7

Writing Papers Model: 1

8

Idea Do Research Write Papers

Writing Papers Model: 2

• Forces us to be clear, focused• Crystallizes what we don’t understand• The purpose of our research is:

–To convey OUR IDEA

9

Idea Do Research Write Papers

Idea Write Papers Do Research

Conveying the Idea

• Here is a problem• It’s an interesting problem• It’s an unsolved problem• Here is my idea• My idea works (details, data)• Here’s how my idea compares to other

people’s approaches

10

Getting ready to write a paper• What type of paper am I writing?

– conference, journal etc.

• What do I wish to state?• Have I got all the background work with

me?• Have I organized the paper?• What electronic format will I use?

– LaTeX, MS Word, …

11

Contents of a paper• Title of the paper• Authors, designations, and addresses• Abstract (150 to 250 words)• Keywords• Sections describing the work

– Introduction (Background)– Previous Work (Literature Survey)– Proposed Work (Algorithm, Design, Methodology)– Analysis (Complexity Analysis, Quantitative

Analysis, Statistical Analysis, …)– Implementation and Results– Conclusions

12

Title

• Should not be too long– “Short and Sweet”

• No Abbreviations or Acronyms– (Unless well established terms like PID)

• Well reflective of your Main Contribution– It should reflect the novel part of your research

• Try to “coin” an attractive (or catchy) title• Should be followed by Names and

Affiliations– According to the Journal’s or Conference’s format

13

Abstract• Usually 1 paragraph, generally between 100-

250 words• Major points to include

– Problem statement/Overview of current research situation/Motivation

– Proposed technique (the main contribution of your paper)– The solution (you should describe the results briefly)– A brief conclusion

Should Not Include:– References– Avoid “general motivation”– The sentences in the abstract should not be repeated

later in the paper “word-for-word”– Avoid equations and math

14

Introduction• Define the problem

– Why is it worth solving?– What did others do?– Why did they not succeed?– Why do I believe I did better?

• Contribution of the paper– Is there anything new in the paper?– How good are our results?– Summary of our approach/results

15

Stanford InfoLab’s Patented 5-Point Structure for Introductions

• What is the problem?• Why is it interesting and important?• Why is it hard? (E.g. why do naive

approaches fail?)• Why hasn’t it been solved before? (Or

what’s wrong with previous proposed solutions? How does mine differ?)

• What are the key components of my approach and results? Also include any specific limitations

16

State Clearly the Contributions

17

Related works

• Brief description of the existing body of work• Citations to published work

– In [2], Rosetti and Longfellow described the meaning of life. Wordsworth presented a different view point in [3].

• Bring out the specific advantage of your work w.r.t. published work.– However, the Rosetti-Longfellow formula [2] fails

for technologies below 0.2 micron. In this paper, we extend their formula to the deep submicron domain.

18

Fallacy: To make my work look good, I have to make other people’s work look bad

Fallacy: To make my work look good, I have to make other people’s work look bad

Related works

• Point out both advantages and disadvantages of related work– (provides context; defuses objections; be honest)

• Discuss all references that you cite

DO NOT• Write a laundry list• Bash the references• Include irrelevant references• Write a paragraph about a very peripheral work

19

Sections and Subsections• Organize each section into subsections

and (possibly) sub-subsections1. Introduction1.1 Problem Description1.1.1 Inductance Extraction1.1.2 Transmission Line Models

20

The proposed Technique

• Can be divided into several sub-sections• Provide a general description of the

technique or method you proposed (use block diagrams where appropriate)

• In another section, start with the general theory first then followed by your own formulations

• Write down your algorithms (equations)/strategy and provide explanation on them

• Figures may be required where necessary

21

Hardware/Software Development

• Provide a description of how the software is developed – Computer specs, Programming Languages used,

Flowcharts, Block diagrams, etc.

• Provide sufficient description of the hardware setup/ designed – technical specifications, block diagrams, sensor

limitations, components used, photos of the hardware, etc.

• Highlight the uniqueness of your hardware compared to similar products (if any)

22

Application of Proposed Technique

• Usually new techniques/algorithms are tested using simulations

• Provide a write-up on how the simulation is carried up – including all assumptions used, how important

parameters are chosen/selected, etc.

• Similarly for hardware application, describe how the experiments have been carried out

• Only discuss experiments which are relevant to show the effectiveness or superiority of the proposed technique

23

Results and Discussions• Results can be tabulated in several forms

– use graphical form where applicable

• Comparison with existing techniques• Provide discussions

– correct analyses of experimental results show the quality of the paper

– Always provide answers to why the experiments went such a way

• Every figure shown in the paper must be addressed in the text

• What conclusion can you draw from your experiments?

24

Figures and Tables

• Number all the figures and tables• Graphs are also shown as figures• Provide captions for all figures and tables

– Fig. 3. Floorplan of the SONAR Chip

• Refer to each figure and table– In Figure 3, we show the floorplan of the

SONAR Chip as obtained using the MASON software

25

Conclusions

• Highlight the major advantages of the technique/method proposed when compared with other techniques

• Don’t just repeat the introduction/abstract• Re-emphasize the novelty of the research

work• Write down the limitations of your research

work if any such that reviewers cannot ‘attack’ the weaknesses of the technique

• Write suggestions to improve the problems or limitations of the proposed technique (such as in future work)

26

References

• References are important in a technical paper

• With adequate references, they reflect that you have made a – substantial study/survey of the existing or

current techniques, – understand their limitations and disadvantages,

in short it means you understand the problem statement

• All references must be cited in the text of your paper

27

Language and Style: Tense• Text can be written in either the past or present

tense• Past tense is OK for describing results of an

experiment but use present tense for a general conclusion– “We observed that the expression of Bmp4 WAS

increased. This suggests that the gene IS regulated by Shh.”

• Present tense is more immediate and indicates that a process is ongoing

28

WHATEVER TENSE IS USED, BE CONSISTENT ANDDON’T SWITCH BACK AND FORTH IN THE SAME

PARAGRAPH !!!

WHATEVER TENSE IS USED, BE CONSISTENT ANDDON’T SWITCH BACK AND FORTH IN THE SAME

PARAGRAPH !!!

Visual Structure

• Give strong visual structure to your paper using– sections and sub-sections– Bullets– Italics– laid-out code– Find out how to draw pictures, and use them

• Paragraphs are important to break the text up into readable units

• The first and last sentence of a paragraph is most important

29

Visual Structure- Example

30

Use the Active Voice

• The passive voice is “respectable” but it DEADENS your paper. Avoid it at all costs

31

Use Simple, Direct Language

32

Frequent Errors Writing a Paper• Avoid excessive use of boring verbs such as

“show, observe, occur, exhibit…..”• Avoid complex ways of saying a simple thing

– “The results showed that dog weight increased” versus “The dogs weighed more”.

• Use of “suggest that ….” ; “hypothesize that….” “possible that….”– These phrases do not need “may”, “might”– e.g. “Our results suggest that Hoxa3 may be

involved in thymus development” (not correct)– “Our results suggest that Hoxa3 is involved in

thymus development” (correct)

33

Before Sending to the Journal

• Have the paper read by several people– Listen to what they say, especially if same

criticism comes up several times

• Check and recheck spelling, figures, references, legends etc – Reviewers can be really annoyed by careless

editing and mistakes reflect badly on your science

• Make sure you have followed all the requirements of the journal about electronic submission etc.

34

Some Strategies in Writing a “Good” Paper

• Identify the novelty of the research work– Study related works and compare your approach– Identify the current problems and limitations– Find how to overcome the limitations

• List down the strengths and weaknesses of the proposed technique– Do not allow reviewers to attack your weaknesses

• Provide a good analysis on the research results– Do not just write down what has been observed

from your experiments– Explain why these observations are ‘so and so’– Provide a summary of results and how results led

to the conclusion

35

Summary

• If you remember nothing else:– Identify your key idea– Make your contributions explicit– Use examples to describe the idea

36

37