Transcript HOW TO HAVE A GOOD PAPER

HOW TO HAVE A GOOD PAPER
Tran Minh Quang
What and Why Do We Write?
Letter
 Proposal
 Report for an assignments
 Research paper
 Thesis
….

Bad Writing




No one know what you are doing
You can not explain what you want to do
Can not get the budget
Can not reach better position for your career
Bad Writing





The topic is so general or so detail
Bad organization
The content is not clear and is disorderly
written
So many typo and lexicological mistakes
Dose not follow the format
Good Writing







The right topic is chosen
Good organized
Follows the format
The content is written in a good logical order
The content is rich enough
Easily to be understood and interpreted
The reference list is carefully cited
The Organization of Good Papers







Topic
Abstraction
Introduction
The related works
The main procedures (including 2 to 5
sections)
Conclusion and future work
Reference list
Choosing Topic







Is clear
Not too general
Not too detail
Not too long
Not to short
Meaningful
Has a nice rhythm
Choosing Topic (con’t)




Topic 1: “Research on Data Mining”
Topic 2: “Studying Association Rules Technique in
Data Mining”
Topic 3: “Association Rules is a Good Technique for
Finding the Customers’ Behavior”
Topic 4: “Applying Association Rules Mining
Technique to Analyze Customer Behavior ”
WHAT IS THE GOOD TOPIC?
Writing the Topic and Subtopic



Capitalized all or the First letters of nonarticle words
Articles with less than 3 letters are written in
lower-case
Example: “Applying Association Rules
Mining Technique to Analyze Customer
Behavior ”
Abstraction




To summarize the paper
After reading the abstract, the reader can
stop reading the paper
Written in only one paragraph
About 100 to 250 words
Abstraction

The abstract includes:
–
–
–
–
–
What have been done
By which methods the research was
conducted
The summary of the results
Brief comparisons with other methods
Dose not include the motivation!!!
Introduction
The motivation of this research
 Methods that are used in the research
 Why those methods are used?
 What/What methods other research
used
 What is the main achievement of the
research
 The organization of the paper

Related Works


What have been done and by what methods
in other researches
What is the main different between this and
other researches
–
–
–
–

Different approaches
Different data set
Different results
Reversional hypotheses
By what means these differences
were reached
Main Procedures
Is the heart of the paper
 About 2 to 5 sections
 Explains the methods and procedures
that the research was conducted
 How to get the data
 How the experimental results can be
achieved

Main Procedures

Other researchers can follow these
procedures to get the same results
Conclusion and Future Work




To conclude what has been done and the
main results achieved
What is the prominent achievement
compared to other researches
What is back-ward of this research that
should be solve in the future
What kind of researches or applied works
should be stemed from this researches
Reference List
Every research has to be stemed from
other works
 References have to be cited in the
content of the paper
 The format of citing reference list is be
long to the regulation of the
organization where the paper is
submitted to

Reference List
[index] <Authors>, <the name of the paper>.
<The proceedings or journal>, <publishers>,
<volume> <year> <page number>
Reference List
References
[1] Agrawal, R, and Srikant, R. Fast algorithm for mining
association rules. In proc. of VLDB ’94. pp. 487-499, Santiago,
Chille, Sept. 1994.
[2] Han, J., Pei, J., and Yin, Y. Mining frequent patterns without
candidate generation. In proc. of ACM SIGMOD Conference on
Management of Data, pp. 1-12, 2000.
[3] Bayard, R.J. Efficiently mining long patterns from databases. In
proc. of ACM SIGMOD Conference on Management of Data,
pp. 85-93, 1998.
[4] Grahne, G., and Zhu, J. High performance mining of maxima
frequent itemsets. In proc. of SIAM’03 workshop on High
Performance Data Mining, 2003.