Eienvalue based feature extraction - PowerPoint PPT Presentation

Title: Eienvalue based feature extraction


1
Research Importance
Dr. D. S. Guru dsg_at_compsci.uni-mysore.ac.in
Department of Studies in Computer
Science, University of Mysore, Manasagangotri, Mys
ore 570 006
2
Research
Search Again
3
(No Transcript)
4
(No Transcript)
5
(No Transcript)
6
(No Transcript)
7
(No Transcript)
8
(No Transcript)
9
Research
Search Again
An Active Process of inquiry

• Systematic
• Diligent

Aims at Discovery
Interpretation Revising facts
Solution to a problem Making practical
applications possible Applied
Greater Knowledge Basic Research Pure/fundamental
10
Research
Pure Basic/foundation Asking and Answering
questions Advances knowledge Greater impact and
general (original)
Applied Application driven Asking and Answering
needs Uses knowledge Solve an application
problem (Need not be original)
May insight into new theoretical work

• No clear separation between pure and applied
• No controversy, at least in researchers mind

11
In any case
Research is Innovative

• It makes improvements
• It introduces something new
• It devices a new method
• It exploits successfully a new idea
• It creates a new dimension of performance

OR Contributory
12
Why should a teacher do?

• builds tomorrow
• has to motivate students
• kindle fire in the young brains
• create vibrations in even in the dull brain

13
A teacher is successful
if he / she can inspire student
To.. think BIG in his / her own
perspective grow BIG in his / her own
pursuit perform BIG in his / her future life
14
Research Category

• Exploratory Research
• (Structures and identifies new problems)
• Constructive Research
• (Develops solutions to a problem)
• Empirical Research
• (Tests the feasibility of a solution)

15
Research Process

• Formulation of the topic
• Motivation, Purpose, Objectives, Scope must be
  clear
• Generate hypothesis
• Inductive knowledge (Assimilation)
• Deductive knowledge (Elimination)
• Establish/prove the hypothesis
• Gathering of data
• Analysis of data
• Theoretical / practical
• Conclusion, revising of hypothesis
• Documentation
• Publishing
• Manual

16
Success in research largely depends on

• Ability of the researcher
• Understanding of the researcher
• Domain knowledge of the researcher
• Commitment of the researcher
• Love and affinity of the researcher towards the
  problem
• (Researcher should get married with the problem)

Then
?
What is the role of the supervisor?
17
Documentation

• Proposals
• Should convey plans
• Abstracts
• Compressed version of accomplishments
• Synopsis
• Extended Abstract
• Research papers for publication
• Transformation of knowledge
• Reports or thesis
• A detailed container

18
Paper publication (way to measure research)

• Conference Proceedings
• No review but presentation
• Reviewed but not critical
• Indexed

• Journal
• (Peer reviewed)
• Ideas and support
• Complete system

19
Characteristics of a publication

• Novelty
• Relevance (Scope)
• Reference to prior work
• Clarity of presentation
• Technical correctness
• Validation
• Impact

20
Paper Format
?Title
?Abstract
?Keywords
1. Introduction
2. Proposed Model
3. Algorithm
4. Experimentation
5. Discussion
6. Conclusion
Acknowledgement
References
21
Paper Format
?Title
?Abstract
?Keywords
1. Introduction
Relevance of the problem
2. Proposed Model
Statement of the problem
3. Algorithm
Review - State of the art
4. Experimentation
Objectives Accomplished
5. Discussion
Organization of the Paper
6. Conclusion
Acknowledgement
References
22
Paper Format
?Title
?Abstract
?Keywords
1. Introduction
2. Proposed Model
New Model
3. Algorithm
Step by Step Design
4. Experimentation
Justification - Proof
5. Discussion
Illustrative Demonstration
6. Conclusion
Acknowledgement
References
23
Paper Format
?Title
?Abstract
?Keywords
1. Introduction
2. Proposed Model
3. Algorithm
Algorithm - Abstract
4. Experimentation
Hand Simulation - if reqd.
5. Discussion
Apriori Analysis
6. Conclusion
Acknowledgement
References
24
Paper Format
?Title
?Abstract
?Keywords
1. Introduction
2. Proposed Model
Experimental Analysis
3. Algorithm
4. Experimentation
Aposteriori Profiling
Case Studies
5. Discussion
6. Conclusion
Acknowledgement
References
25
Paper Format
?Title
?Abstract
?Keywords
1. Introduction
2. Proposed Model
3. Algorithm
4. Experimentation
5. Discussion
Comparative Analysis
Quantitative
6. Conclusion
Acknowledgement
Qualitative
References
26
Paper Format
?Title
?Abstract
?Keywords
1. Introduction
2. Proposed Model
3. Algorithm
4. Experimentation
5. Discussion
Summary
6. Conclusion
Gist of Accomplishments
Acknowledgement
Scope for further research
References
27
Thesis Writing
Abstract
1. Introduction
2. Contributory Chapters
3. Comparative Analysis
4. Future Work
5. Complete System
6. Conclusion
Authors publications
Appendix
Bibliography
28
Thesis Writing
Abstract
1. Introduction
Preamble
2. Contributory Chapters
Statement of the problem
3. Comparative Analysis
Background
4. Future Work
Motivation
5. Complete System
State of the art
6. Conclusion
Methodologies Proposed
Authors publications
Results Accomplished
Appendix
Organization of the Thesis
Bibliography
29
Thesis Writing
Abstract
1. Introduction
2. Contributory Chapters
Each chapter should be self contained independent
contribution
3. Comparative Analysis
4. Future Work
Each chapter should be in the format of a
research paper
5. Complete System
Chapters should be arranged in Hierarchical/Evolut
ionary sequence
6. Conclusion
Authors publications
Appendix
Bibliography
30
Thesis Writing
Abstract
1. Introduction
2. Contributory Chapters
When more than one method is proposed to solve a
problem then the performance indices of all the
methods should be compared
3. Comparative Analysis
4. Future Work
5. Complete System
6. Conclusion
Authors publications
Appendix
Bibliography
31
Thesis Writing
Abstract
1. Introduction
2. Contributory Chapters
3. Comparative Analysis
Different perspective for quality improvement
4. Future Work
5. Complete System
Supportive/ Extensional
6. Conclusion
Exploitation for possible applications
Authors publications
Appendix
Bibliography
32
Thesis Writing
Abstract
1. Introduction
2. Contributory Chapters
3. Comparative Analysis
4. Future Work
Platform
5. Complete System
System Features
6. Conclusion
Working principles
Authors publications
Appendix
Bibliography
33
Thesis Writing
Abstract
1. Introduction
2. Contributory Chapters
3. Comparative Analysis
4. Future Work
A review summary
5. Complete System
Overall contributions
6. Conclusion
Observations, limitations and Recommendations
Authors publications
New avenues for further research
Appendix
Bibliography
34

• Unfortunately
• The part that is good is not original, and the
  part that is original is not good.
• - Samuel Johnson

35

• After review of the new manuscript, I wrote
• Dr Smith
• we are happy to accept your superbly written
  paper for publication.
• However, I just couldnt help adding tell me
  who wrote it for you.
• Dr. Smith answered
• I am so happy that you found my paper
  acceptable, but tell me who read it for you.
• - Robert A Day ( Editor)

This confusion is always there because,
participating co-researchers vary from paper to
paper.
36
Summary

• Research?
• Types of Research
• Importance
• Categories
• Process
• Role of Researchers and Supervisor
• Documentation

37
R E S E A R C H ?? Reading a lot for
Establishing Scientific and Engineering
Aptitude to have a good personal Rapport with a
Commitment to build up a Healthy society for
the development of Nation -D.S. Guru
No(w) Questions!?
38
Thank you one and all
Dr. D S G