Elements of Scientific Writing: Preparing for Top Conference and Journal Publications
Instructor: Prof. Julian Cheng Date: August 24 (Saturday) & 25 (Sunday), 2019 Time: 9:00-12:00 & 14:00-18:00 Location: CI-101, Tsinghua SIGS, Shenzhen
Abstract
Scientific writing for both conferences and journals is essential for research-based graduate students. In the current practice, a starting young researcher often learns how to write a scientific paper by mimicking authors of published papers, which can be misleading. His supervisor then laboriously, and perhaps even painfully, corrects the first manuscript draft. Both parties can end up with extreme frustration. Despite the importance of scientific writing, few universities have such a formal course that is specially designed for top conference and journal publications. The purpose of this course is to fill this gap. In this short course (16 lecture hours), assuming that the students have intermediate-level of English language skills, we will cover the following elements of writing a scientific paper: essential units of a research paper and how to write abstract, introduction, methods, results and discussion; common English usage mistakes to avoid; effective writing principles; how to construct strong proses; how to write correct mathematical expressions and illustrate scientific results. We will also cover important topics such as how to communicate with the editors, as well as ethics associated with research publications.
Instructor's biography
Julian Cheng received his B. Eng. degree in electrical engineering from University of Victoria, Victoria, BC, Canada, M.Sc. (Eng.) degree in Mathematics and Engineering from Queen’s University, Kingston, ON, Ph.D. degree in electrical engineering from University of Alberta, Edmonton, AB, Canada. He is currently a Full Professor in the School of Engineering, Faculty of Applied Science, at The University of British Columbia (Okanagan campus) in Kelowna, BC, Canada. His current research interests include wireless communication theory, wireless networks, optical wireless communications, wireless locations, and quantum communications. Dr. Cheng has successfully trained a large number of graduate students on how to write high quality research articles, and many of these students are international students with little prior knowledge on scientific writing. He has co-authored more than 270 journals and conference publications, and vast majority of these publications have appeared in top venues. He now serves as an Area Editor for IEEE Transactions on Communications. In the past, he served as an Associate Editor for IEEE Transactions on Communications, IEEE Transactions on Wireless Communications, IEEE Communications Letters, and IEEE Access, and was a past Guest Editor for a special issue of IEEE Journal on Selected Areas in Communications on optical wireless communications. Currently, he serves as the head of the electrical engineering program in the School of Engineering at UBC as well as the President of the Canadian Society of Information Theory.
Lecture 1: Introduction to Scientific Writing
Different kinds of research publication
Research report, technical report
Thesis dissertations
Research monograph (research orientated books)
Conference versus journal publications
Difference between technical writing, academic writing, and scientific writing.
Elements of a scientific paper
Peer Reviewed never publish in non-peer-viewed journals.
Results Reproducible never publish papers with incorrect results.
Create New Knowledge some journals publish survey papers.
Permanently Archived available to information retrieval services.
IMRD(Introduction Method Result Discussion) Structure
-
Titile
- Title should be concise and define the scope of the paper.
-
Abstract
Abstract contains a summary of the paper, a brief description of the problem, the solution, and conclusions. Be concise, e.g., TCOM allows Abstract with only 200 words. Avoid reference citation and abbreviations.
How to write an Abstract? Draft your Abstract by picking out major objective/hypotheses and conclusions from your Introduction and Conclusion sections, select key sentences and phrases from your Method section, reveal your finding from your Results section. Link your sentences and form them into a single paragraph and re-write. Good writing is all about re-writing.
-
Keywords
- IEEE papers typically require five keywords listed in alphabetical order.
-
Introduction
Don’t underestimate the importance of Introduction.
A good introduction tells a good story and cite the related papers in the course of doing so.
Provide motivations of the entire paper. Review and cite all relevant papers.
Provide a brief description of the research problem to be addressed and the proposed solutions.
Tip: include a paragraph starting with a sentence that reads “The contributions of this paper are as follows.”
Provide an organizational overview of the entire paper. It can be omitted due to space constraints.
-
System Model
Write this part first
Define every symbol immediately after its first occurence
Establish good notations
Define every symbol immediately after its first occurrence
-
Methodology and Solutions
Core part of the paper and should be worked out before you start writing any paper. Foundation of the entire paper.
New results are not the grounds for publications.
Writing is not about communicating to the reader about your idea. It is about changing their ideas.
Tip: Work out special cases and link your special results to the ones that have already been published.
A paper with novelty is likely to be well received. You want to give the reviewers an Aha moment.
A paper with elegant and useful analytical results is always worth publishing.
A paper that is correct and explains a counter-intuitive phenomenon is always worth publishing.
-
Numerical Results & Discussions
For certain analytical results, it is always good idea to verify using Monte Carlo computer simulations.
The parameter values should be taken, whenever possible, from systems in the real world.
All numerical should be interpreted. You need to explain why curves look the way they do.
It is generally good idea to compare your numerical results with the ones from a comparable system in existing literature.
Tip: Font size for the figures should be sufficiently large so that they are still readable after being scaled down.
-
Conclusions
- Conclusions summarize the work and highlight the most significant findings. This is also where the future work should be discussed.
-
Acknowledgement (optional)
If an anonymous reviewer or Editor provides constructive remarks that helped improve the presentation of the work, thank him generously.
Thank any individuals who provide helpful comments. Tip: it is always a good idea to send the manuscript to your trusted colleagues for inputs before submission, provided they are available and willing to do so.
Don’t forget to thank your funding agency.
-
Appendix (if any)
- As a rule, any lengthy tedious mathematical derivations that can be skipped without compromising the understanding of the paper should be moved into the Appendix or Appendices.
-
References
Reference contains a list of papers that have been cited in the paper. They should be listed in the order of citation.
Whenever possible cite the journal version, which is more readily available.
Never cite an unpublished work.
In general, it is not a good idea to cite a webpage.
See IEEE Editorial Style Manual (available from www.ieee.org/documents/stylemanual.pdf).
How to improve your writing?
Learn how to write strong proses. (Grab a copy of Style: Lessons in Clarity and Grace by Joseph M. Williams and The Elements of Style by William Strunk Jr. and E. B. White)
Practice, practice, and practice.
You can learn any subject by writing the subject down using your own words. (Read Writing to Learn by William Zinsser.)
Learn how to edit other people’s papers.
Learn from good IEEE papers.
Double Submission (NO!)
- 投稿，被要求大修后，不能换期刊投，要换也得先withdraw，但是不推荐这么做，因为这是逃避的表现
How to avoid plagiarism
Present your original ideas in an original way
Always cite all related works and give credits to others
Learn how to document quotation and cite properly
Always cite your own related work
-
Use a plagiarism checker software: WriteCheck and Turnitin
Lecture 2: Ethics in Research Publications
A Researcher’s Reputation is Everything
What is the difference between research and plagiarism?
➢The act is considered to be plagiarism if you use one source for your point of view within your research paper.
➢But it is considered research if you use more than one source to cover the topic.
Types of plagiarism with severity levels
➢Paraphrasing plagiarism (Severe)
➢Verbatim plagiarism (Serious)
➢Global plagiarism (Severe)
➢Using other’s work and claiming your work, including paying for essays or papers.
➢Many these “customized papers” are plagiarised and have poor quality.
➢If caught, the student can either fail the course or get expelled by his/her university.
➢Universities can also revoke degrees.
➢Self-plagiarism ( Moderate)
This policy is found in the IEEE Policies document, Sections 6.4.1B(f) and 6.4.1B(h). These two sections are given below.
(f) Plagiarism is unacceptable. The verbatim copying or reuse of one's own research) as indicated in paragraph "h" below) is considered another form of plagiarism or self- plagiarism; it is unacceptable.
(h) Except as indicated in Section 6.3.4 (Multiple Publication of Original Technical Material in IEEE Periodicals), authors should only submit original work that has neither appeared elsewhere for publication, nor which is under review for another refereed publication. If authors have used their own previously published work(s) as a basis for a new submission, they are required to cite the previous work(s) and very briefly indicate how the new submission offers substantial novel contributions beyond those of the previously published work(s).
➢Incorrect citation (Moderate)
➢Non-existent source (Severe)
Question: Can you submit a journal paper based on a conference paper?
Answer: Consult your individual journal!
➢IEEE Communications Letters /IEEE Wireless Communications Letters: No
➢IEEE Transactions on Communications or IEEE Transactions on Wireless Communications: Yes, but the authors must cite the conference version and state the novelties and contributions beyond the conference papers.
➢ IEEE Transactions on Circuits and Systems: Yes, but 50% additional technical material is expected.
What is double submission? 一稿多投
Concurrent submission of the same paper to different journals
➢Two different papers but with large overlaps can also be considered as double submission!
➢Double submission can occur without being self- aware.
➢Always withdraw form one journal first before submitting to another journal.
➢If not sure, consult a trusted senior colleague.
How to avoid plagiarism?
➢Present your original ideas in an original way. ➢Always cite all related works and give credits to others. ➢Learn how to document quotation and cite properly. ➢Always cite your own related work.
➢Use a plagiarism checker software: WriteCheck (http://en.writecheck.com/)and Turnitin (www.turnitin.com)
Other forms of author misconduct
➢Submitting a paper without the permission of all co-authors
➢Providing incorrect email addresses of co-authors
➢Extra authors in the final paper submission
➢Reduce the number of pages in the camera-ready version to avoid overlength charge
➢Adding/removing references from the camera-ready version
➢Citing remotely related or total irrelevant papers by the authors or forming a citation social network group
➢Fail to disclose related papers by the authors
Lecture 3: Common Grammar and Usage Errors
Use Precise Words
Use Simple Words
Avoid misused words
Omit Unnecessary Words and Phrases
Replace unnecessary phrases with simpler ones
Avoid Better A considered number of Many An example of this is the fact that For example As a consequence of Because Despite the fact Although During the time that While First of all First Has the capability of Can In many cases Often In some cases Sometimes Considerable amount of Much Based on the fact that Because In order to to Dangling Modifier
Comma splice
➢Jammal wanted to go to the ball game, his friend Jason wanted to see the new movie.
➢Jammal wanted to go to the ball game. His friend Jason wanted to see the new movie.
We have fixed the comma splice by using two separate sentence.
➢Jammal wanted to go to the ball game; (on the other hand), his friend Jason wanted to see the new movie.
We have fixed the comma splice by using semicolon instead of comma.
➢Jammal wanted to go to the ball game, but his friend Jason wanted to see the new movie.
We fix the comma splice by using a conjunction word (for and, nor, but, or, yet, so) , also known as FANBOYS, after the comma.
Don’t start a sentence with a number
➢(32) can be derived from ...➢Equation (32) can be derived from ...
The above style rule applies when the equation number occurs in the first position in a paragraph. Otherwise, use ``Eq. (32) can be derived from ...”
➢[10] is concerned about ...➢Reference [10] is concerned about ...
Use “Denote” correctly
➢The color red is used to denote passion or danger.
➢The dashed line in Figure 1 denotes profits.
➢N denotes the set of natural numbers.
➢We denote by P(X) the set whose elements are the subsets of X.
➢We denote the set whose elements are the subsets of X by P(X).
➢The set whose elements are the subsets of X is denoted by P(X).
➢We denote N as the set of natural numbers.➢We denote the set of natural numbers by N.
“Minimum” or “Minimal”?
Minimum: smallest, least
Minimal: small amount of
➢The minimum bet for this Blackjack table is $10.
➢The function attains its minimum (value) when x=0.
➢There is a minimal risk to invest in IBM stock.
➢When the housing market crashes, you can buy this house at minimal cost.
Remarks: In math, “minimal value” means “local minimum”.
Similarly, maximum means largest, and maximal means large amount of.
Minima is the plural of minimum, and maxima is the plural of maximum.
“Electric” vs. “Electrical” vs. “Electronic”
“Electric” refers to devices that run on electricity or produces electricity, and it can be used figuratively.
It is acceptable to say an electric kettle or an electric tooth brush or electric cars, or the atmosphere is electric.
“Electrical” is concerned with electricity.
It is acceptable to say electrical engineers, electrical engineering. Normally we don’t say electric engineer or electric engineering.
“Electronic” describes the devices with internal electrical components that manipulate electrical currents; It also refers to virtual form of things that exist in the traditional world
electronic circuit board, electronic textbook, electronic file
“Relationship” or “Relation”?
The word “relation” is used in formal writing, and it is used to describe the connection between large group of people, like countries, companies, shareholders, etc.
Examples are diplomatic relations, industrial relations, public relations, and race relations.
➢A company needs to improve its relations with its shareholders.
➢He does not want to endanger US trade and other relations with China.
“Relation” sometimes means extended family member.
➢I am going to visit my friends and relations. ➢James Thompson was in court today accusing Cindy Thompson, no relation, of stealing his rabbit.”
We use “relation” to describe how two things are connected.
➢An inverse relation between wind speed and density fluctuations appears to exist.
➢This observation led us to hypothesize that there may exist a relation between the structural properties of the antenna array and the capacity of the channel.
We use ``relationship” to describe the connection between smaller groups of people.
➢He had very strong relationships with elected officials.
➢We have strong relationships with our European suppliers.
➢But perhaps most importantly, a mother and daughter relationship is permanent.
You can use ``relationship” to describe a situation with a (romantic) partner.
➢She was in a relationship with the Governor of California. ➢There were strong relationships in her life, but no marriages.
You can use relationship to describe how two things are connected. In this case, you can say ``relationship between ... and ...”or relationship between two ...”
➢The mechanism implied in this relationship between the two processes may be, at least in this study, an explanation for the general finding that the tropopause is a transport barrier.
➢There exists a relationship between the storm's intensity and genesis location in the different phases of ENSO along with turbulent sea surface temperatures in the Northwest Pacific Ocean
Conclusion: In scientific writing, you can use either “relation” or “relationship” interchangeably to describe how two things are connected. But for the other situations, relation cannot be replaced with ``relationship” and visa versa.
Lecture 4: Principles of Effective Writing – Characters and Actions
Your sentence should tell stories about characters and their actions.
A sentence is clear when its important actions are in verbs.
Readers want actions in verbs, but even more they want characters as subjects.
Pasive voice or negative voice?
Actions they performed as part of their research, actions that anyone can perform: measure, record, examine, etc.
➢The depth was measured...
➢The speed was recorded...
➢Three types of modulations are employed in ....
➢The BER at the receiver is measured using ...
➢A clear bump at N=3 is observed in Fig. 5.
Actions refer to the writers’ own writing or thinking: cite, show, inquire, argue, claim, etc.*
➢We claim ...
➢We argue ...
➢We observed...
➢We have chosen the value of that is optimum when receiver noise dominates the multiple-access noise...
➢For purposes of analysis, we assume that the true transformed pulse shape is known to the receiver....
➢We will first discuss the iterative construction of local neighborhoods, which will provide a basis for our derivation.
Lecture 5: Principles of Effective Writing – Stress and Emphasis
Put familiar/simple information at front
When readers build up momentum in the first nine or ten words of a sentence, they more easily get through complicated material that follows.
➢The derivation of a method for calculating error probabilities resulting from the combination of additive random noise with lSI and timing error is the subject of this study.➢In this paper, we derived a method for calculating error probabilities resulting from the combination of additive random noise with lSI and timing error.
In general, your sentences should begin with elements that are relatively short. A short introductory phrase or clause, followed by a short, concrete subject, followed by a verb expressing a specific action.
➢The investigation of the calculation of error probabilities for ISI channels using Fourier series appears in the last part of this section.➢In the last part of this section, we investigate the calculation of error probabilities for lSI channels using Fourier series.
The general principle is to carry the reader from simplicity to complexity.
Put complex/new Information near the stress point
The last few words of a sentence are also particularly important, because they receive special emphasis.
We’ll call this most emphatic part of a sentence its STRESS.
➢Core losses of magnetic materials in electrical engineering are decomposed into three components, according to Bertotti’s losses separation model.➢According to Bertotti’s losses separation model, core losses of magnetic materials in electrical engineering are decomposed into three components.
Just as we look at the first few words of a sentence for point of view, we look to the last few words for emphasis.
Three Tactical Revisions
- Trim the end.
➢ The DIRMA receiver is based on the theory of hypothesis testing in the way that a hypothesis is proposed and tested on the recorded data.➢ The DIRMA receiver is based on the theory of hypothesis testing.
- Shift peripheral ideas to the left.
➢ The data offered to prove ESP are weak, for the most part.➢ For the most part, the data offered to prove ESP are weak.
➢ Job opportunities in computer programming are getting scarcer, and it must be remembered.➢ It must be remembered that job opportunities in computer programming are getting scarcer.
- Shift new information to the right.
➢ Questions about the correction of the turbulence contaminated bits are more difficult [than something just mentioned].➢ More difficult [than something just mentioned] are questions about the correction of the turbulence contaminated bits.
You can easily check whether you have stressed the right words by reading your sentences aloud.
Lecture 6: Principles of Effective Writing – Concision
Six principles for concision:
-
Delete meaningless words.
kind of actually particular really certain various virtually individual basically generally given practically
-
Delete words that repeat the meaning of other words.
full and complete, true and accurate, hopes and desires, hope and trust, each and every , first and foremost, any and all , basic and fundamental, various and sundry
-
Delete words implied by other words.
Redundant Categories
large in size, unusual in nature, of a bright color, round in shape, of a strange type, at an early time, honest in character, area of mathematics, in a confused state
General Implications
encoding the information source.source coding.信源编码
Replace a phrase with a word.
Change negatives to affirmatives.
Delete useless adjectives and adverbs.
Lecture 7: Principles of Effective Writing – Shape
Break a long sentence into several short sentences.
Long and complex, but also clear and shapely
Rule
Get the subject as quickly as possible
Get to the verb and object quickly
Avoid interrupting the subject-verb connection
Avoid interrupting the verb-object connection
Get into your main point as early as possible
Avoid adding one subordinate clause to another to another
Lecture 8: Principles of Effective Writing – Elegance
Parrelel structure - easy to read
Writing with parallel structure involves presenting related thoughts in similar fashion or the same parts of speech.
Ways to end a sentence with special emphasis:
-
Weighty Words
Nominalizations > Nouns > Adjectives and Adverbs > Prepositions
Of + Weighty Word
Lecture 9: Principles of Effective Writing – Cohesive and Coherence
Cohesive: a sense of flow
Coherence: a sense of whole
The core logic: put old information before new information
Readers expect to see familiar information in the beginning of a sentence
Readers expect new information at the end of the sentence
Cohesive versus Coherent
coherent tells a story, but cohesive just combine things
-
examples of cohesive but not coherent:
流水账
unrelated subjects
no common themes
no central sentences
No distraction at the beginning of the sentence
Make subjects topics
Lecture 10: Effective Punctuations in Scientific Writing
Lecture 11: Writing Mathematics in Scientific Papers
Math writing requires slow reading
expersses complex ideas
must be read and pondered several times
must be read selectively and in pieces
General rules
write clear and professional
formulas and equations follows grammatical rules
Avoid use "I", but use "We", even though you are the only author of the paper
Write in the active voice and present tense
Do not put symbol at the beginning of the sentence
The structure of a proof
present outline and the key ideas
-
keep the read informed of where you are in the proof
The aim/idea is to ...
Our first goal is to show that ...
Now for the harder part ...
The only, but crucial use of ... is that ...
Give a reference if you omit the proof
-
Reference format:
For example, Theorem 5-1[cite here]
by the First Fundamental Theorem (if the theorem has its name)
When you end the proof, write "now the proof is complete"
Too many definitions may overwhelm a reader
Minimize the distance of the definition and its first use place
Respect fundamental constant
- For instance, i = 根号1, pie, e
Notational conventions
Lower case Greek letters represent scalars
Lower case roman letters represent column vectors
Upper case represent Matrix
Simplify the expressions
An equation is displayed when
It needs to be numbered
It would be hard to read if placed in-line
It merits special attention
For instance, it contains the first occurrence of an important variable
Too long formula should be broken before a binary operation (e.g., +, −,×)
Parallelism should be used to aid readability
Mathematical expressions are part of the sentence
Don't use mathematical symbols unless they serve a purpose
- unless the name is used in a following sentence
Watch out for unnecessary parentheses
- As in the phrase “the matrix − is singular”
Avoid starting a sentence with a mathematical expression
BAD: “ is an ill-conditioned matrix”
GOOD: “The matrix is an ill-conditioned”
BAD: “ represents the minimum distance”
GOOD: “Let denotes the minimum distance”
Separate mathematical symbols by punctuation marks or words
Bad: If>1 <0.
Fair: If>1, <0.
Good: If > 1, then < 0.
BAD: ..., 1 ≤ ≤ 2
FAIR: ... for 1 ≤ ≤ 2
GOOD: ... (1 ≤ ≤ 2)
The use of the article “the” is inappropriate when the object to which it refers is not unique or does not exist.
- Change it to 'a'
Refer to theorems by their numbers or names
Bad: The previous theorem, the proposition above
Good: By Section 3.4 and Theorem 5.6, the proposition holds
Good: The sampling theorem states that a signal of bandwidth can be reconstructed
Add a word or phrase when you reference an earlier equation
Use something more informative than just “equation” or “Eq.”
Bad: “From (6.2)”
Good: "From the definition (6.2) of dual norm”
Bad: “Combining (3.14) and (2.9)”
Good: “Combining the recurrence (3.14) with the inequality (2.9)”
Figures and tables should be labeled.
- They should be referred to in the text (before they appear)
When citing a book or article, include a theorem number or page number
- Omit this only in cases in which you really mean to cite the entire work
Referencing and Citation (cont’d)
When citing an arXiv preprint, include the version number or precise date
When citing a preprint elsewhere on the web, give the URL and date of the manuscript or date downloaded
Cite “forthcoming work” only if there is a publicly available preprint
Not recommended for citing a website
Using Figures
Use visual arguments in your mathematical writing
A good graph should convey relevant and specific information
Use math programs like Maple and Matlab to generate smooth figures
Fully explain how a picture fits into your mathematical argument
Bad: Looking at the graph, we can see that the result is true.
Good: The graph increases sharply at = 3, confirming our earlier prediction that the robots will begin a homicidal rampage three years from now.
Some More Guidelines
Don't use a dummy variable if it is already being used for another purpose
Note the difference between the Greek letter epsilon, , and the “belongs to” symbol ∈
Note that the distinction between the Greek letter and the product symbol （连乘符号）
-
Standard mathematical functions such as sin, cos , tan, log , min, max , gcd, trace, and lim are set in roman type
- It is a common mistake to set these in italic type
-
Bracketing multilayered expressions
- have a choice of brackets for the layers as well as sizes
Write “the th term”, not “the ^th term”, “the ’th term” or “the -th term”
-
A slashed exponent, as in 1/2, is generally preferable to a s
1 tacked one, as in fraction style
-
The standard way to express that is to take the values 1 to in steps of 1 is to write
- = 1, ⋯ , or = 1, 2, ⋯ ,
-
Avoid tall inline expressions
- but use [g1, g2]^T
-
There are two different kinds of ellipsis
-
Vertically centred ⋯ (\cdots in Latex)
Is used between operators such as +, =, ≤
Examples: + +⋯+ , ≥ ≥⋯≥ 12312
-
Ground level ... (\ldots in Latex)
Is used between a list of symbols or to indicate product
Examples: 1,2,...,, 12 ...
-
Use a vertically centred dot for denoting multiplication in expressions
Particularly, where terms need to be separated for clarity Examples:
16046641 = 13 ∙ 37 ∙73 ∙457
-
-
Indicate the type of a symbol when using it in a sentence
Bad: For Each and ∈ , + ∈
Good: If and are any two points on the elliptic curve , then their sum + is also a point on
Symbols in different formulas should be separated by words
-
Choose names for indices and ranges consistent with the usual ordering of the alphabet
Bad: The index ranges from 1 to and ranges from 1 to
Good: The index ranges from 1 to and ranges from 1 to
-
Use suggestive/mnemonic notation.
- Examples: for set, for function, for ball, etc
Tricks and Pitfalls
-
Choose a target audience level of expertise/background
- 看人下菜，文章读者的文化程度，决定了文章要写多深
Use counterexamples to clarify the limitations of the analysis
Make proper use of “very,” “trivial,” “easy,” “nice,” “fundamental”
Use grammar checking, spell checking, and plagiarism detection platforms, e.g., Grammarly
-
Write an ordered tuple in (...) , instead of {...}
Example: a sequence should be written
Bad:{} ≥0
Good: () ≥0
-
The prefix “non” does not need a hyphen（连字符-） in common words such as
- nonempty, nonnegative, nonsingular, nontrivial, and nonzero
-
Filler words like “We remark that” and “Note that” should be omitted,
- Except when their absence would produce a sentence starting with a symbol -
If a proof ends with an end-of-proof symbol, then it is not necessary to write “This concludes the proof.”
-
Don't use the two words which and that interchangeably
The word which is used to introduce an idea or qualifier that isn’t necessary to understand the sentence
Bad: Groups which have prime order are cyclic
Good: Groups that have prime order are cyclic
Good: Navy beat Army 18-14, which is good news
-
One of the meanings of the word respectively is in the order given
Use it for defining several things at once
Good: Let and be the centers of circles and , respectively
Capitalize when referring to a particular theorem, e.g., This proves Theorem 7
A Mathematical Writing Checklist
-
Is your paper neatly typed?
-
Make sure that you have included all of the diagrams and
graphs
Make sure that the paper is double-spaced and has wide enough margins
-
-
Has the paper been proofread?
Sloppy work is not appreciated. Do check over everything
-
Is there an introduction?
Make sure that you explain the problem to the reader
Give some indication of what will follow in the paper
-
Did you state all of your assumptions?
Write down any physical assumptions that you made
Write down any mathematical assumptions that you made
-
Is the mathematics correct?
- This should be obvious
-
Are the grammar, spelling, and punctuation correct? Is the writing clear and easy to understand?
Make sure that there are no sentence fragments
Equations and formulas should have correct punctuation
Make sure that your paper flows smoothly
Don’t be careless! Check your spelling!
-
Are all of the variables defined and described adequately?
Make sure that you introduce each variable that you use
Describe each variable as precisely as possible
Don’t forget any units!
-
Are the mathematical symbols used correctly?
Don’t use an “=” sign outside of a formula
Make sure that the symbols are not misused
Use equations and formulas where they are appropriate
-
Are the words used correctly and precisely?
- Avoid using vague language and too many pronouns
-
Are the diagrams, tables, graphs, and any other pictures you include clearly labeled?
Use computer to draw smooth Graphs
Indicate the scales in the labels
Glossary for Mathematical Writing☺
Without loss of generality, i.e., I have done an easy case
By a straightforward computation, i.e., I lost my notes
The details are left to the reader, i.e., I can't do it
The following alternative proof of 's result may be of interest, i.e., I cannot understand
It will be observed that, i.e., I hope you hadn't noticed that
Correct to within an order of magnitude, i.e., Wrong
An exercise on technical writing
References
[1] Knuth, D.E., Larrabee, T., Roberts, P.M. and Roberts, P.M., Mathematical Writing, Mathematical Association of America, 1989
[2] Krantz, S.G., A primer of mathematical writing, American Mathematical Society, 2017.
[3] Nicholas J. Higham, Handbook of Writing for the Mathematical Sciences, Society for Industrial and Applied Mathematics, 1998.
[4] The Chicago manual of style, 16th edition, University Of Chicago Press, 2010.
[5] James Kilpatrick, If we could only get this one right, Eugene Register- Guard, Jan. 2007 .
[6] Douglas B. West, The grammar according to West, Sept. 2015. Available at http://www. math.illinois.edu/~dwest/grammar.html .
[7] Dimitri Bertsekas, Ten Simple Rules For Mathematical Writing, MIT, April 2002.
[8] Kevin P. Lee, “A Guide to Writing Mathematics.”
[9] B. Poonen, “PRACTICAL Suggestions for Mathematical Writing.”
[10] Clarke, D.J., Waywood, A. and Stephens, M., “Probing the structure of mathematical writing.” Educational studies in mathematics, 25(3), pp.235-250, 1993.
Lecture 12: Figures and Tables
In academic works, visual contents matter!
➢Objective of an academic work is to disseminate the scholarly findings to the readership of a wide spectrum.
➢Readership often determines the worthiness of an academic work by observing the visual evidence.
➢ Use Meaningful Presentation with Few Possible Illustrations of Strong Impact
What is Meaningful Presentation?
➢Each figure and table should able to tell a story or meaningful research outcome without going through another illustration.
➢The story of a figure and table should be understood immediately by making a quick observation.
De-facto Rules of Meaningful Presentation
➢Figures and tables should be consistent with the description of the text.
➢Legends of the figures and titles of the tables need to stand on their own.
➢Figures and tables need to be numbered in a same order they appear in the text.
Placement of Illustrations: Where and How?
➢Important rule: Place the figure and table in the place where it is expected.
➢Always keep the readership in mind to select a certain type of illustration.
Figure or Table: How to Select Appropriate Form of Presentation?
➢Use figure when pattern is important.
➢Use table when the actual value matters.
➢Use textual description when more precise description is required to compare a few data points.
Number of Illustrations vs. Impact of the Illustrations
➢Keep the illustrations simple and few as much as possible to tell one complete story.
➢Also Measure the impact of the illustrations. ✓Avoid figures of repeated observation.
Producing Professional Figures: Options
➢Microsoft Excel can convert experimental data and export to several professional graphic software.
➢Adobe Illustrator, Adobe Photoshop, ChemDraw, Microsoft Visio can be used for preparing system model/block diagram.
Important Notes on Professional Figures
➢Lines should have good contrast against a white background.
➢Axes, curves, and lettering must be solid and smooth.
➢All the components should be appropriately labeled.
➢Cluttering needs to be avoided.
不要仅用颜色区分线条，黑白打印版看不出来区别
Producing Professional Figures for Final Submission
➢It is good idea to draft/prepare figures at an early stage.
➢Most of the IEEE Journals prefer figures either in EPS or TIFF format.
➢Prepare the figures according to the recommended formats.
Clarity and Integrity for Producing Figures
➢The presented data in a figure can sometimes be exaggerated or misleading.
➢Care needs to be taken to make the figures as accurate as possible to illustrate the actual facts.
How to Avoid Misleading Figures
➢Provide the relevant information of each figure to make it reproducible.
a) The exact parameters used in the experiment.
b) The number of observations or sample size.
c) Any other statistical information.
How to Avoid Misleading of Figures (Cont’d)
➢Be aware about the limitation(s) of the data presented in a figure.
a) State all the necessary assumptions.
-
b) When comparing two or more sets of data,
please state that under what condition such a comparison is valid.
Types of Figures
➢The figures presented in an academic work can be either photograph, drawing and diagram, and graphs.
➢The selection of specific type of figure depends on the objective.
Guidelines of Using Photographs in the Scholarly Works
➢Purpose: To illustrate the realism of the observed data in an experiment or an experimental system setup.
➢Key rule: Always aim to make photographic images as informative as possible.
Important Instructions for Photographs
➢Label everything that is relevant.
➢Use scale bars in the micrographs.
➢If a record is not a obvious in the image, draw a separate diagram to emphasize the presence of such a record.
➢Check whether color photograph is acceptable by the intended venue of publication.
Purpose of Block Diagram
➢Block diagrams are particularly important in the engineering academic papers.
➢Objective:
➢To illustrate how certain things, such as electrical, mechanical, or chemical objects, react with each other.
Guidelines of Block Diagram
a) Keep the number of blocks small.
b) Use consistent and meaningful name for each block.
c) Use arrows to indicate the direction.
d) Follow the same direction showed in the diagram in the textual description.
e) Do not make the block diagram crowded.
Guidelines for Graphs
✓Type of common graphs in an academic paper
a) Line graph.
b) Scatter plot.
c) Bar graph and Pier chart
d) Box plot.
✓It is important to know how to present data by an appropriate graph.
Line Graph: Purpose
➢Most popular type of graphs in the scientific papers.
✓It is Primarily used to make comparisons, often over time.
✓Example: How the capacity of a wireless communication channel varies with respect to the transmission power.
Line Graph: Guidelines
➢Three or four curves should be maximum in a line graph.
➢Individual curves need to recognized properly, especially when they cross.
Scatter Plot: Purpose and Generation
➢Purpose: Find correlation between two sets of data obtained from an experiment.
➢MATLAB function, such as, scatter, scatter3, are useful to produce scattered plots.
Bar Graph: Purpose and Generation
➢Purpose of bar graph:
➢Compare the findings for different instances of an experiment when there is no continuity among the findings.
➢Use Vertical bars of same width in a bar graph. ➢MATLAB function, such as, bar, bar3, are useful to produce bar graph.
Box Plot: Purpose and Generation
➢Purpose: Provide statistical characteristics of a given set of data.
➢Box plot can be either horizontal or vertical
➢MATLAB function, boxplot, can be used to generate box plot from a given set of experimental data.
Four Key Rules of Selecting Appropriate Graph
a) Use line graph for dynamic comparison.
b) Use scatter plot to find a correlation for a collection of data.
c) Use bar graphs when findings can be subdivided and compared.
d) Use pie chart to compare the parts of a whole.
Table: When and How to Produce?
➢Purpose of the tables: ➢Present the facts more concisely by enabling side-by-side comparison.
➢Repetition of same data in the figures, tables, and text should be avoided.
➢Professional table can be produced by using Microsoft Excel and Microsoft word.
Placement of Contents in a Table
➢Place the most familiar item on the left side, and new information on the right side.
➢Organize the contents in a logical manner. ➢Listings and numbers in a table should be aligned.
➢Similar components (such as, hours in day) are preferred to be arranged vertically.
How to Deal with Large Tables?
➢Oversized table should be avoided; however, sometimes they are inevitable.
- a) Try to reduce the number of unnecessary columns.
- b) Try to breakdown a large table into two or multiple small tables.
Guideline for an Algorithm
➢What is an algorithm?
➢An algorithm is a sequence of well-defined and finite instructions required to complete a specific task.
➢It is visually much easier to comprehend if the algorithm is written in the form of a pseudocode.
Structure of a Pseudocode
a) Title.
b) Input variables.
c) Initialization of the required parameters.
-
d) List of the control operations with the help of
repeat, for, while, if/else.
e) Output: The variables that contain decision obtained by the algorithm.
伪代码每行的行号要标上
Lecture 13: Abstract and Title
The title (followed by the abstract) is the first thing a reader sees, and so it should be the last thing an author writes (just after the abstract).
An abstract is a brief summary which condenses in itself the argument and all the essential information of a paper.
An Abstract needs to be dense with information but also readable, well-organized, brief, and self- contained.
The purpose of a title and abstract is often described as marketing tools for the paper:
get someone who has read the title read the abstract
get someone who has read the abstract read the paper.
How long should an abstract be?
Short articles 85-150 words
Journal articles: 75-250 words
Conference abstract: 150-1000 words
Master’s Thesis: 150-300 words
Ph.D. Dissertation: 350 words
Due to word limitation, the abstract must be concisely written.
The contents in an Abstract
The abstract should be a concise, stand-alone summary of the paper that covers the following topics:
Background/motivation/context Aim/objective(s)/problem statement Approach/method(s)/procedure(s)/materials Results
Conclusion(s)/implications
Two types of abstract
Conventional abstract （EE/CS领域用这种）
Structured abstract
Conventional Abstract
In the conventional abstract, the five basic components are not identified with explicit headings
Conventional Abstract is more popular among science and engineering disciplines
Guidelines in writing an abstract
Written in less technical way with the most significant results
Try to avoid acronyms（缩写）, because the abstract is written for a broad audience.
There is no reason to define acronyms if you don’t plan to use them in the abstract. （后面要用到这个词，才需要缩写节省篇幅）
Abstract shouldn’t cite any references.
Abstract should contain minimum math equation or math symbols.
Try to avoid first person and choose verb tense properly.
Research gap/problem uses simple tense
• The main problem, however, is ... • We examine why these models have difficulty with ... • However, this assumption is not valid when ... • This is complicated by ... • However, this assessment cannot be based solely on ... • Although it is known theoretically that ..
Methodology uses past simple tense or simple tense
• Two catalysts were examined in order to ... • Sample were prepared for electron microscopy using ... • The effect of pH was investigated by means of ... • The data obtained were evaluated using ... • Numerical examples are analyzed in detail ... • The calculated wavelengths are compared to ... • Several methods are created using ... • The accuracy is evaluated by ... • A detailed comparison is made between ...
Results use simple tense or past simple tense or both
• The model consistently underestimates ... • The ratio shifts towards .... • The most accurate readings are obtained from ... • These example illustrates that ... • The hydrocarbons showed a marked increase in .... • This finding correlated with ... • The experiments demonstrated there are two matrices ... • It was found that proteins are produced from ...
Achievements use present perfect tense
• We have obtained accurate quantitative LIF measurement ... • This investigation has revealed that ... • We have devised a strategy which allows ... • We have demonstrated the feasibility of this approach by ... • A novel material has been produced which ... • Considerable insight has been gained concerning ...
Applications use present simple tense
• This process is suitable for production of ... • This framework can be used to evaluate ... • This approach can be applied to ... • These profiles may serve as a predictor for ... • This framework can be used to evaluate
Applications use present simple tense
• This process is suitable for production of ... • This framework can be used to evaluate ... • This approach can be applied to ... • These profiles may serve as a predictor for ... • This framework can be used to evaluate
Dissecting conventional abstracts(long article)
The speed of sound in a fluid is determined by, and therefore an indicator of, the thermodynamic properties of that fluid.
The writer provides background factual information, which is usually derived from the beginning of the Introduction
The aim of this study was to investigate the use of an ultrasonic cell to determine crude oil properties, in particular oil density.
The writer combines the method, the general aim and the specific aim of the study in one sentence.
An ultrasonic cell was constructed to measure the speed of sound and tested in a crude oil sample. The speed of sound was measured at temperature between 260 and 411 K at pressure up to 87 MPs.
The writer summarizes the methodology and provides details. How much details given here will largely depend on the emphasis of the paper. If the contribution is the methodology, provide some numerical details.
如果论文偏实验性质，那么可以在摘要部分写具体环境，比如温度气压，这是novelty体现的地方，但如果论文偏理论，那么最好不要说具体数据
The measurements were shown to lead to an accurate determination of the bubble point of the oil.
The writer indicates the achievement or finding of the study.
这里应该量化一下有多accurate，因为这是一个主观的指标
This indicates that there is a possibility of obtaining fluid density from sound speed measurements and suggests that it is possible to measure sound absorption with an ultrasonic cell to determine oil viscosity.
The writer presents the implication of the study.
Dissecting conventional abstracts(short article)
This paper reports the use of a novel water-soluble polymer lend as a coating to control drug release.
The writer combines what the paper does (This paper reports), the method or materials used (water-solutable polymer blend), the contribution (novel) and the aim of the study (to control drug release)
短文摘要要非常精简，多思考如何用一句话表达更多信息
It was found that using a blend of methylcellulose and a water-soluble copolymer significantly slowed the release rate of ibuprofen compounds in vitra and allowed for a more consistent release rate of 10-20% per hour
The writer refers to the method in more detail and provides numerical details of the results.
Structured abstracts（especially in medical journal)
A structured abstract contains the following five headings:
-
Background:
What issues led to this work?
What is the environment that makes this work interesting or important?
-
Aims or Objectives:
What did you plan to achieve in this work?
What gap is being filled?
-
Approaches or Methods:
How did you set about achieving your aims (e.g., experimental method, simulation approach, theoretical approach)?
What did you do?
-
Results:
- What were the main results of the study (including numbers, if appropriate)?
-
Conclusions:
What were your main conclusions?
Why are the results important?
Have you achieved your objectives?
Where will they lead (future work)?
Benefits of structured abstracts
It forces the author to include information from all five categories.
It makes these five sections easy to find and access.
Creating a conventional abstract from a structured abstract systematically
Write a structured abstract using appropriate headings
When you are finished and satisfied with the result, simply delete the headings and combine all of the lines into one paragraph.
Finally, reread this new abstract and change the beginnings of sentences to increase readability and flow, if needed (though this will usually not be necessary).
Outcome: a well-written and effective abstract with most of the benefits of a structured abstract.
相当于去掉heading的、经过cohesive美化后的conventional abstract
好处：条理清晰、不会缺少五元素、很快产出
How to make a title?
The title is the single most important phrase of a paper.
Guideline in preparing a title
Attractreaders
Informative and Accurate: State the main topic of your research
Make your paper outstanding from the other competing articles
Title example
Effect of Hormones on Tumor Cells
Weak title because ``what hormones” and “which tumor cells”? Title is too vague and unspecific.
Effect of Testosterone and Estradiol on the Growth of and Morphology of Rat Epithelial Tumor Cells
Title example
The Kinetic Analysis of the Na-ATPase in the Corneal Epithelium
Articles at the beginning of a title can be omitted.
Kinetic Analysis of the Na-ATPase in the Corneal Epithelium
General tips for making titles
Title is normally a phrase, which should not be overly long.
A complete sentence with active verb is not a good title.
Try to populate your title using the keywords.
Unless universally known abbreviation (e.g. DNA, AIDS, UFO) or chemical compound, avoid abbreviation in a title.
In general, avoid title with question, unless when there is no simple answer can be provided.
Avoid word such as ``Novel” in the title.
Final check on your title -- Answer the following question:
Would someone interested in the exact topic of your paper, reading this title, be inclined to read the abstract?
Introduction
When to write the introduction?
- In practice, you will find that you need to be certain about what you have done and what you have found in order to write the Introduction, and so the best time to write it will be after you have written, or at least drafted, the report sections.
Two language areas which are important in the Introduction:
Tense pairs
Passive/Active
Passive/Active (we have discussed before)
Use Active voice when it refers to your own thoughts/assumption/...
Use Passive voice to achieve cohesion/coherence.
一般我们在Introduction引用各种参考文献，有个很好的方法来分辨引用是否生硬，那就是把引用叉掉，句子还读的通，句子之间仍然有连贯性，那么这样的Introduction就写的很好，作者就是在说故事
Build a model for the introduction**
Introduction
1 Polylactide (PLA) has received much attention in recent years due to its biodegradable properties, which offer important economic benefits. 2 PLA is a polymer obtained from corn and is produced by the polymerisation of lactide. 3 It has many possible uses in the biomedical field [1] and has also been investigated as a potential engineering material [2], [3]. 4 However, it has been found to be too weak under impact to be used commercially [4].
5 One way to toughen polymers is to incorporate a layer of rubber particles [5] and there has been extensive research regarding the rubber modification of PLA. 6 For example, Penney et al. showed that PLA composites could be prepared using blending techniques [6] and more recently, Hillier established the toughness of such composites [7]. 7 However, although the effect of the rubber particles on the mechanical properties of copolymer systems was demonstrated over two years ago [8], little attention has been paid to the selection of an appropriate rubber component.
8 The present paper presents a set of criteria for selecting such a component. On the basis of these criteria it then describes the preparation of a set of polymer blends using PLA and a hydro-carbon rubber(PI). 10 Th is combination of two mechanistically distinct polymerisations formed a novel copolymer in which the incorporation of PI significantly increased flexibility.
Analyze:
Begin the introduction
- Most research articles begin by indicating that the research field or topic is useful or significant.
Sentence 1
“Polylactide (PLA) has received much attention in recent years due to its biodegradable properties, which offer important economic benefits.”
The writer establishes the importance of this research topic.
刚开始不要写的很深奥，要slow
Begin the introduction
- Research articles often begin with accepted or established facts. This ensures that the reader shares the same level of background information as the writer, and is therefore ready to read the article.
Sentence 2
‘PLA is a polymer obtained from corn and is produced by the polymerisation of lactide.’
The writer provides general background information for the reader.
Begin the introduction
- Significance and Background
Sentence 3
‘PLA has many possible uses in the biomedical field [1] and has also been investigated as a potential engineering material [2], [3].’
The writer does the same as in Sentences 1 and 2, but in a more specific/detailed way, using research references to support both the background facts and the claim for significance.
General problem area
Sentence 4
“However, it has been found to be too weak under impact to be used commercially [4].”
The writer describes the general problem area or the current research focus of the field.
- Remember to keep this general description of the problem area or current research focus brief.
Transition between the general problem area and the literature review
Sentence 5
“One way to toughen polymers is to incorporate a layer of rubber particles.”
The writer provides a transition between the general problem area and the literature review.
Literature review
Sentence 6
“For example, Penney et al. showed that PLA composites could be prepared using blending techniques [6] and more recently, Hillier [7] established the toughness of such composites.”
The writer provides a brief overview of key research projects in this area.
As a general rule, you should include references to previous or current research wherever it is useful, even in a sentence whose function is primarily to provide a transition.
You can’t just ‘pour’ the literature review onto the page in any order;
You should arrange your references and studies so that the reader is able to process them in a logical way.
Introduce your research purpose Sentence 7
“However, although the effect of the rubber particles on the mechanical properties of copolymer systems was demonstrated over two years ago [8], little attention has been paid to the selection of an appropriate rubber component.”
The writer describes a gap in the research.
This is where you begin to introduce the purpose of your paper and the specific problem you will deal with, and in order to do this
it is necessary to create a research space.
Introduce your research purpose
It is conventional to introduce your research purpose with a signaling connector such as However or Although. Don’t be shy about pointing out the problems in previous research.
It may be necessary in order to explain why you have done your study.
The language used here is usually respectful and impersonal and is therefore not considered offensive.
Introduce your research purpose
Sentence 8
“The present paper presents a set of criteria for selecting such a component.”
The writer describes the paper itself.
You can describe it, say what its purpose or focus is, give its structure or a combination of these. You normally use the Present Simple tense to describe the work itself (This paper is organized as follows... or This study focuses on...)
Give the methodology and finding of your paper
Sentence 9
“On the basis of these criteria it then describes the preparation of a set of polymer blends using PLA and a hydrocarbon rubber(PI).”
The writer gives details about the methodology reported in the paper.
Sentence 10
‘This combination of two mechanistically distinct polymerisations formed a novel copolymer in which the incorporation of PI significantly increased flexibility.’
The writer announces the findings.
Give the methodology and finding of your paper
Although you can give information about your methodology or findings in the Introduction, be careful not to go into too much detail at this point or you will find that you have nothing to write about in the Methodology or Results sections.
Summary
1 Establish the importance or your field Provide background facts/information Define the terminology in the title/key words Present the problem area/current research focus 2 Previous and/or current research and contributions 3 Locate a gap in the research Describe the problem you will address Present a prediction to be tested 4 Describe the present paper Lecture 14: Result, Discussion, and Conclusion
Why do we need numerical results?
One of the most important reasons for writing a Results Section is that you must communicate your own understanding and interpretation of the results to your readers.
We need numerical results to validate analysis, verify our experiments, demonstrate the usefulness of our proposed algorithm, compare the proposed scheme with other existing schemes
Build the model for numerical results
You can begin the numerical results by describing the scenarios of your numerical experiments and re-emphasize the system model assumptions.
You must specify your parameters of your simulations/experiments clearly and precisely.
-
compare your results with other research (in various conditions)
- You need to describe individual results in some detail, selecting results which are important, typical, or especially interesting.
-
Discuss problems found in the results. Don’t ignore problems in your results unless you are certain that the problems are insignificant and invisible.
- Including a discussion of a problem in your work shows that you are fully in control of your research and able to evaluate it clearly. Furthermore, it provides you with an essential element for the Discussion/Conclusion: directions or suggestions for future research.
Summary
1 Describe the scenarios of your numerical experiments; Re-emphasize the system model assumptions. 2 Specify your parameters. 3 Specific/key results in detail, with explanations; Comparisons with results in other research; Comparison/s with model predictions. 4 Problems with results; Possible implications of results. Discussion and Conclusion
Some journals end with a subsection titled Discussion, some end with a subsection titled Results and Discussion and others end with a subsection titled Conclusions.
Here we focus on the Conclusions.
Build the model for Conclusion
-
Revisiting previous section
- Summarizing/revisiting general or key results
-
Mapping
- Relationship to existing research
-
Conclude your main findings/contributions
- Summarize your main findings/contributions. Refine your implications.
-
Indicate the possible applications
What if my work doesn’t have any obvious applications?
You can look at the beginning of your Introduction, and the first sentences and paragraphs of related work in your field to find the possible application of your work.
Many research studies don’t have obvious applications. You don’t need to search for or try to create applications where there are none..
如果审稿人的要求你没法达到，你可以把达不到的要求放在conclusion里，就说这些达不到的要求是future work :P
abstract和conclusion最大的区别：
abstract是给那些没看过你文章的人写的，所以要吸引
conclusion是给那些看过你文章的人写的，所以要总结
Summary for Conclusion
1 Revisit previous sections; Summarize/revisit key results. 2 Mapping (relationship to existing research). 3 Findings/contribution; Refine the implications. 4 Limitations; Current and future work; Applications Lecture 16: Effective Communication with Editor and Reviewers
Professor Cheng's own experience!
How to assess a manuscript and write a review report?
以下是Prof. Cheng自己做审稿人的经验，知己知彼:P
Peer review is about striving towards the TRUTH – the very quest of scientific enquiry!
Paper versus Manuscript
before publishment is manuscript
after publishment is paper
Peer reviews can benefit your own career! Here are the benefits
Gaining first-hand experience of the key considerations that go into the publication decision
Keeping up to date with novel research in your field
Having an opportunity to demonstrate your expertise
Types of peer review
Single-blind peer review 单盲，作者不知道审稿人，审稿人知道作者
Double-blind peer review 双盲，双方都不知道对方是谁，比如IEEE Communication Letter
Open peer review 公开，谁都可以comment，审稿人可酌情考虑public opinion，这种要求很高
Before you begin
Do I have the appropriate expertise to review this manuscript?
Do I have any conflict of interest regarding this article or its author?
Do I have the time to conduct this review effectively?
Managing your review
Treat the materials you receive as confidential documents • Criticize the work, not the authors
Read journal guidelines carefully
Adapt your review to suit the journal’s requirements
Write clearly and precisely
Read 1st time: Gaining an overview
The first read-through is a skim-read
Aim to form an overall impression and understanding
Make a first draft describing the main aim of the manuscript
Try to bear in mind the following questions
What is the main question addressed by the research?
How original is the topic?
Is the manuscript well written?
Are the conclusions consistent with the evidence and arguments presented?
Flag major flaws of the manuscript Examples:
Presenting results contrary to the author's original claim
The use of a discredited method
-
Ignoring a process that is known to have a strong
influence on the area under study
Concluding the First Reading
Draft the first two paragraphs of your review
The first one should summarize the research questions addressed by the manuscript
The second one should present the contribution of the work
The first paragraph should:
-
Help the editor properly contextualize the research and
add weight to your judgement
-
Show the author what key messages are conveyed to
the reader
Focus on successful aspects of the manuscript
The second paragraph should cover the following points
Is the manuscript's premise interesting and important?
Are the methods used appropriate?
Do the data support the conclusions?
Read 2nd time: The science
The second read-through is a detailed review
You should identify:
Any places where the meaning is unclear or ambiguous
Any factual errors
Any invalid arguments
You may also consider the following aspects of the article
The article’s contribution to the discipline
Academic rigour and accuracy
Style and structure
One purpose of the second read to help prepare the manuscript for publication.
You may still decide to reject it following a second reading.
Offer clear suggestions for how the authors can address the concerns raised
Separate your points into “Major” or “Minor” issues
Distinguish between
Correctable versus uncorrectable problems • Major versus minor concerns
How to write an author's reply?
Key points (Principles of author)
Be professional and courteous
Make it easy for the reviewers and the editor
-
Your reply should be self-contained document
蓝字or红字把改动的地方标出来，还要写Major Changes
让审稿人感到舒服，For your convenience, ...
-
Keep in mind that your reply will be read by all the reviewers and the editor
- 不仅是回给审稿人A，审稿人B、C，甚至编委也会来看
-
Practice the principle of positive reinforcement （Prof. Cheng's trick)
comment everything reviewers wrote, especiallly praise
因为有时审稿人自己也会忘掉自己说的好话
其他审稿人也会看到，之前的审稿人原来赞同这篇文章
用同样的单词，重复一遍
Return the exact wording back to the reviewers
The authors’ reply is your opportunity to impress the reviewers and the editor
-
Never argue with the reviewers, but you can still disagree with them in a diplomatic way
如果审稿人说你的文章和某篇文章很像
你可以说：感谢您的attention，帮我们找到一篇遗漏的reference :)
第一步要agreable with reviewer's opinion，语气基调必须让审稿人“爽”，感谢您的批评blahblah，这个reference我们不知道blahblah
第二步条理清晰地用道理说为什么不一样
-
第三步，必须要做点什么 (Action)
- To address reviewer A's concern, in the revised introduction, we will discuss the detailed difference between these two paper （比较两篇文章的相似点和不同处）. Furthermore, we will paste our revision in this responding copy&paste revised sentence(把修改后的句子粘贴在邮件里面)
-
Do point-to-point reply and avoid points-to-points reply
- 不要同时回复多条意见，审稿人会觉得你想略过去
The reply has no page limit. Keep a high writing standard
If major revision is allowed
Carefully study editor and reviewers’ comments
-
Do additional research
(more analysis and experiments, more data collection, more graphs, more comparison)
Revise the manuscript accordingly • Be carful about the page limit of the journal
Prepare a point-to-point author’s response
Resubmit to the same journal within the specified time frame
Contact the editorial staff if more time is required for preparing the revisions
If resubmission doesn’t happen by the deadline, most journal will consider the manuscript rejected automatically
If rejection, what next?
It’s not the end of the world!
Don’t submit to another journal without revisions • Some researchers will frown at you
Always revise and improve the manuscript before resubmitting to another journal.
Basic elements in replying every comment
-
Acknowledgment
Acknowledge the reviewer’s comment on suggestion
The tone should be in agreement
-
Discussion
This is the place where you can offer some
disagreements with the reviewer with very detailed and convincing argument
-
Action
- What are the action taken to address reviewers’ comments?
-
Quotation
- For the convenience of the reviewer, quote the revised text in italic
Answering typical questions from the reviewers
Scenario 1 The paper is not novel and similar ideas have been done by the other authors
既然审稿人没说具体是哪篇文章，那他有可能说的是这篇文章，在这篇文章里blahblah（分析异同点
主动权在自己手上hhh
Scenario 2 Reviewers require many changes that you can only address 75% of them
The Role of Editors
Role of Editors
Gate keeper of the journal （不要接收烂文章
Keep in mind Editor makes the decision not the reviewers （审稿人的意见只是参考
Can editor rejects a manuscript if the majority of reviewers provide positive reviews? Yes!
Take a cursory（adj.草草的）need of the manuscript and get an initial impression the manuscript
Identify the expert reviewers if transaction paper or invite generalist in the case of a survey paper
Honor authors (no reviewer request 作者可以要求不要哪些审稿人
Occasionally, you want to purposefully invite the no list paper if you have very strong reason to do so
Editor can always rescind the reviewers’ comments
Keep a list of good reviewers
Be familiar with the names of good exemplary reviewers
-
Don’t afraid to invite big guys if they are still active and can
provide prompt review
-
Some researches will only accept a review invitation if he
can complete the review within a week
What makes a good editor
Find competent reviewers and make sure all reviewers submit on time
Write a good editorial decision letter
Makes the decision in a timely manner
Be a good gate keeper
Accept every manuscripts or the majority of the manuscripts
Being impartial, e.g., favor some authors on friends
-
Focus on the merits of the paper and take out the human
factors
Be a friend to reject paper co-authored by big guys
Don’t follow up wide reviewers
Don’t communicate with authors on inquiries
Slow to make decision
Decision with no or very few justifications
-
Accept a paper that is either incorrect or a paper with
marginal contributions and poor English
Possible editorial decisions
Accepted on the first submissions (almost never occurs )
Accepted with pending revisions (JSAC)
Major revisions
Rejection with resubmission allowed
Some IEEE journals allow one more resubmission)
Rejected with no resubmission allowed