How To Start Research: V. Jayalakshmi

How to start research
V. Jayalakshmi
Why do we research?
To solve a problem
To satisfy an itch To gain more market share/ Develop and improve
It is part of the Job

To make more money
To get a Ph.D Degree
What are the types of Academic Research Works?

A totally new innovation or frame work Any analysis/ approach/ new algorithm / procedure with at least some deviation from existing methods so as to have contribution which can reduce cost/time/ and or increase in performance
How to proceed?
DECIDING ON A TOPIC
What am I most interested in? Do I want to know a lot about a little or a
little about a lot?
Narrower papers are easier, broader ones more difficult
AREA OF INTEREST
Pick a Direction or area of interest Based on our background/ knowledge/ Exposure (Subjects handled, Courses taken, readings Conferences attended seminars etc.)
Example - Networking
This is an area only. It could be Multicasting in Networks Could be further refined to 'Congestion control for Reliable multicast Further, 'Congestion control through Admission Control
Do not say, Wireless Networks (too general) it can be Wireless with infrastructure / Ad-hoc
Networks / Wireless Sensor Networks

Narrow down further, Systematic Testing of Wireless MAC layer/ Efficient hand off for IP mobility / Security by modifying the protocol in any of the above networks
Do not just say Network Security

but, Security through
Protocol
Encryption/ decryption algorithms Key management
Hardware deployment
Do not say as Quality of Service in
networks, but as, Quality of Service

in IP Networks through Congestion Control
In unicast In Multicast
KEY WORDS
Compile a set of Key Words on the chosen topics Start searching for the key words from
IEEE library on line ACM library on line
Google (selectively)
Citeseer.com
Searching for papers in General area of Computer Networks, look for

IEEE/ACM Transactions on Networking IEEE Journal on Selected Areas of communications Computer Network Journal (Elsevier) And many others
On Wireless Networks ..
ACM Mobicom ACM MobiHoc ACM MANET Journal AdHoc Networks Journal (Elsevier) IEEE / ACM IPSN
IEEE WCNC
And many more Journals and conferences
SELECT PAPERS
From the topics of papers we have collected, Select papers that we think most related to what we have in our mind Do not read all the papers now Check the title, abstract, authors and affiliations, Journal or Conference
Refine Key words ..

Now try to refine our key words and perform multiple searches for quality works
If a specific paper is liked, look at the list of references this will give a good indication to follow
Consolidate
With these additions narrow down to 15- 20 papers
of most related, we think, for the selected area of work
Gather Information
Now start reading ONLY THE ABSTRACT, INTRODUCTION and CONCLUSION Skim the rest of the paper for general idea Identify the emphasis of each paper
Pick the best

Based on our reading and understanding pick 4-6 papers that We think that address our research interests and challenges in the field most appropriately Read these 4-6 papers thoroughly from the beginning to the end
Analyze
Main approaches Methods of Analysis Metrics
Evaluation tools
Analysis and interpretation of results Simulation or Measured data Conclusions
Some questions on these ideas..

Did all/some papers use similar approaches? Have they used same evaluation criteria? Same method of analysis?
If not,
What are the strengths / weaknesses of each method?
Keep a list of ideas we want to explore This will create another list of readings
Keep a list of
What do we think the authors may have missed
The gaps and Limitations that could

be improved upon
Any idea of how to accomplish these improvements
Proposal
Write two pages of proposal defining as clearly as possible, the following:
Motivation Research Challenges Overview of the existing work Limitations of the existing work Potential directions and ideas for improvement
Expected results and impact in the field
Ask the trust-worthy friends to read and comment on these two pages about If it is not clear, go back and rewrite If we think we have missed some other work, go back to the 15-20 list and pick another 3-5 good papers to read and rewrite the proposal
Focus on an idea
Be selective in readings Try to focus, not get confused
Write these two pages

The next step is to write a 10 page proposal elaborating on the 2 pages
Publish
Send paper to Journal/Conference
Do not mind if not accepted
This will give good comments by referees Improve upon the paper utilizing the comments It will get accepted and published
Thesis and Synopsis

If 2 Journal papers and 2 conference papers published, think of consolidating the work Left out works, links that may be needed have to be
completed
Write the draft Thesis Write the Synopsis
Thank You!

How To Start Research: V. Jayalakshmi

Uploaded by

Copyright:

Available Formats

How To Start Research: V. Jayalakshmi

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

How To Start Research: V. Jayalakshmi

Uploaded by

Copyright:

Available Formats

How to start research

To satisfy an itch To gain more market share/ Develop and improve

It is part of the Job

To get a Ph.D Degree

What are the types of Academic Research Works?

Narrower papers are easier, broader ones more difficult

Networks / Wireless Sensor Networks

Do not just say Network Security

Encryption/ decryption algorithms Key management

Do not say as Quality of Service in

networks, but as, Quality of Service

Searching for papers in General area of Computer Networks, look for

Refine Key words ..

of most related, we think, for the selected area of work

Pick the best

Some questions on these ideas..

What do we think the authors may have missed

The gaps and Limitations that could

Any idea of how to accomplish these improvements

Expected results and impact in the field

Write these two pages

Thesis and Synopsis

You might also like