Tema 6.- Publicación de resultados científicos

Transcripción

Tema 6.Publicación de resultados científicos
Introducción
¿Cómo leer un artículo científico?
¿Cómo escribir un artículo científico?
¿Cómo presentar un artículo científico?
Herramientas de soporte

¾
¾
¾
Latex
Lyx
Gnuplot
Redes Inalámbricas y Computación Ubicua/2006-2007
Introducción

¾
¾
¾
Latex
Lyx
Gnuplot
Redes Inalámbricas
Inalámbricas yy Computación
Computación Ubicua/2006-2007
Ubicua/2006-2007
Redes
¿Porqué leer?

Para conocer el estado del arte en un área de interés

Para evitar reinventar la rueda
¾

Ocurre con frecuencia
Encontrar temas de investigación interesantes
Redes Inalámbricas
Ubicua/2006-2007
Redes
¿Porqué no leer?

No se pueden leer todos los artículos

No se debe leer todo

Puede limitar o eliminar la innovación
¾
Una vez conocidas soluciones a un problema en particular, es difícil pensar
de un modo distinto
Redes Inalámbricas
Ubicua/2006-2007
Redes
¿Leer o no leer?

Leer, evidentemente

Saber lo que es importante y lo que no

Saber lo que se puede ignorar sin pérdida significativa de
información
Redes Inalámbricas
Ubicua/2006-2007
Redes
¿Qué leer?

Artículos publicados en congresos importantes
¾

Informes técnicos de grupos de investigación importantes
¾

Los journals van unos años retrasados, pero aún así pueden ser muy útiles
Hay que saber qué grupos buscar
Artículos tipo survey (repaso)
¾
¾
¾
¾
ACM Computing Surveys
CACM, IEEE Computer, Spectrum
Más técnicos - IEEE Personal Communications, …
newsletters - ACM SIGCOMM, ACM SIGMOBILE, ...
Redes Inalámbricas
Ubicua/2006-2007
Redes
Sabe por qué quieres leer el paper

Para saber qué se está haciendo (ej., análisis de actas de
congresos)
¾

Trabajos en tu misma área de investigación
¾
¾

Título, autores, resumen
Introducción, motivación, descripción de la solución, resumen,
conclusiones
A veces es útil conocer más detalles, pero no siempre
Trabajos que quieres mejorar
¾
Leer todo el paper detalladamente
Redes Inalámbricas
Ubicua/2006-2007
Redes
En qué hay que fijarse

Autores y grupo de investigación
¾

Solución encontrada a grandes rasgos
¾

Hay que saber dónde buscar un paper sobre un tema en particular
Debe ser posible volver a analizar al paper si hace falta más información
Metología utilizada para la evaluación
Principales fallos
Redes Inalámbricas
Ubicua/2006-2007
Redes
Un trabajo se ha publicado.

Duda, sé escéptico

Si parece demasiado bueno para ser verdad, normalmente lo
es…
Introducción

¾
¾
¾
Latex
Lyx
Gnuplot
Redes Inalámbricas
Ubicua/2006-2007
Redes
Haz a los demás lo que te gustaría que te hicieran a ti
Redes Inalámbricas
Ubicua/2006-2007
Redes
Estructura típica de un artículo:

Title
Author, affiliation, e-mail, …
Abstract
Introduction
¾
¾

Motivation
Problem description
Related work
Solution
...
Performance Analysis
Conclusions
Future Work
References
Redes Inalámbricas
Ubicua/2006-2007
Redes
¿Cuando escribir?
Cuando
tienes resultados realmente expcionales:
P==NP
¾ Casi dá igual el estilo utilizado, lo leerán de todas formas
¾

La mayor parte de los papers no son excepcionales
Escibir correctamente puede marcar la diferencia
¾ Es mejor decir poco de forma clara, que mucho de forma menos clara
¾
Redes Inalámbricas
Ubicua/2006-2007
Redes
Readability
If
the paper is not readable, author has not given
writing sufficient thought
Two kinds of referees
¾ If I cannot understand the paper, it is the writer’s fault
¾ If I cannot understand the paper, I cannot reject it
Don’t
take chances. Write the paper well.
Badly
written papers typically do not get read
Redes Inalámbricas
Ubicua/2006-2007
Redes
Do not irritate the reader

Avoid Using Too Many Acronyms
¾

AUTMA ?!
You may know the acronyms well. Do not assume that the
reader does (or cares to)
Redes Inalámbricas
Ubicua/2006-2007
Redes
How to write a theory paper

Unreadability is not the same as formalism

Reader should be able to understand contributions without
reading all details

If some proofs are not too important, relegate them to an
appendix
¾
Proofs are not as worthy as new proof techniques
Redes Inalámbricas
Ubicua/2006-2007
Redes
How to write a systems paper

Provide sufficient information to allow people to reproduce
your results
¾
¾
¾
people may want to reproduce exciting results
do not assume this won’t happen to your paper
besides, referees expect the information

Do not provide wrong information

Sometimes hard to provide all details in available space
¾
¾
¾
may be forced to omit some information
judge what is most essential to the experiments
cite a tech report for more information
Redes Inalámbricas
Ubicua/2006-2007
Redes
Discuss related work

Explain how your work relates to state of the art

Discuss relevant past work by other people too

Remember, they may be reviewing your paper.
¾
¾

Avoid: The scheme presented by Vaidya performs terribly
Prefer: The scheme by Vaidya does not perform as well in scenario X as it
does in scenario Y
Avoid offending people, unless you must
Redes Inalámbricas
Ubicua/2006-2007
Redes
Tell them your shortcomings

If your ideas do not work well in some interesting scenarios, tell
the reader

People appreciate a balanced presentation
Redes Inalámbricas
Ubicua/2006-2007
Redes
How to write weak results

If results are not that great, come up with better ones

Do not hide weak results behind bad writing
¾

If you must publish: write well, but may have to go to secondbest conference
¾
¾
¾

Be sure to explain why results are weaker than you expected
Only a few conferences in any area are worth publishing in
Too many papers in poor conferences bad for your reputation
Just because a conference is “IEEE” or “ACM” or “International” does not
mean it is any good
If results not good enough for a decent conference, rethink your
problem/solution
Redes Inalámbricas
Ubicua/2006-2007
Redes
Miscellaneous

Read some well-written papers
¾
award-winning papers from conferences

Avoid long sentences

If you have nothing to say, say nothing
¾
¾
don’t feel obliged to fill up space with useless text
if you must fill all available space, use more line spacing, greater
margins, bigger font, bigger figures, anything but drivel
Redes Inalámbricas
Ubicua/2006-2007
Redes
Technical reports

Useful to get early feedback from other researchers

Puts a timestamp on your work

Can include more information / results than might fit in a paper
Introducción

¾
¾
¾
Latex
Lyx
Gnuplot
Redes Inalámbricas
Ubicua/2006-2007
Redes
How to present a paper (at a conference)
Objectives, in decreasing order of importance

Keep people awake and attentive
¾
¾
¾

everything has been tried: play fiddle, cartoons, jokes
in most cases, extreme measures should not be needed
humor can help
Get the problem definition across
¾
people in audience may not be working on your problem
Redes Inalámbricas
Ubicua/2006-2007
Redes
How to present a paper (at a conference)
Objectives … in decreasing order of importance

Explain your general approach
¾

most productive use of your time
Dirty details
¾
¾
most people in the audience probably do not care
a typical conference includes 30+ paper presentations, yours could be
the N-th
Redes Inalámbricas
Ubicua/2006-2007
Redes
Talk outline or not ?

Useful when several ideas discussed in a single talk

Short talks : Skip the outline

Long talks : Include an outline

Make the outline interesting
Redes Inalámbricas
Ubicua/2006-2007
Redes
Text
You want people to (quickly) read your slides

Use big enough font

Do not put too much on one slide
¾

don’t want to keep them busy reading, instead of listening
Use good color schemes
Not blue on yellow
Redes Inalámbricas
Ubicua/2006-2007
Redes
Text

Slide text need not be grammatically accurate

Keep it short
¾
¾
OK to omit some details
fill them in when you present the paper
Practice makes perfect
versus
Practice can improve your presentations
Redes Inalámbricas
Ubicua/2006-2007
Redes
PowerPoint, but not excessively

Everybody has used PowerPoint

No one is impressed by fancy backgrounds anymore

Avoid using gratuitous animation

Standard PowerPoint layouts can be useful
¾
decent font sizes and color schemes
Redes Inalámbricas
Ubicua/2006-2007
Redes
Picture is worth 1000 words

Use illustrations to explain complex algorithms

Omit minor details, focus on the important

They can read the paper to know the exact algorithm
Redes Inalámbricas
Ubicua/2006-2007
Redes
Short talks

May not have enough time to discuss all ideas clearly

Focus talk on one or two ideas

Summarize rest briefly

Better to explain one idea well, than many ideas poorly
Redes Inalámbricas
Ubicua/2006-2007
Redes
How to present a paper

Avoid blocking the screen

Point to the screen, rather than the slide on the projector
Redes Inalámbricas
Ubicua/2006-2007
Redes
How many slides?

Depends on personal style

Rules of thumb
¾
¾
1 slides for 1-2 minutes
Know your pace

I tend to make more slides than I might need, and skip the notso-important ones dynamically

Anticipate technical questions, and prepare explanatory slides
Redes Inalámbricas
Ubicua/2006-2007
Redes
How to present a paper

Practice makes perfect (or tolerable)

May need several trials to fit your talk to available time
particularly if you are not an experienced speaker
Redes Inalámbricas
Ubicua/2006-2007
Redes
If English is your second language

Accent may not be easy to understand

Talk slowly
¾
Easier said than done
Redes Inalámbricas
Ubicua/2006-2007
Redes
No substitute for experience

Nothing like a terrible presentation to learn what not to do

Try to learn from other people’s mistakes, instead of waiting for
your own
Redes Inalámbricas
Ubicua/2006-2007
Redes
Summary

Use common sense

Learn from experience

Enjoy!
Papers can be fun
Introducción

¾
¾
¾
Latex
Lyx
Gnuplot
Redes Inalámbricas
Ubicua/2006-2007
Redes
What is LaTeX?

Typesetting Language
Designed for Science and Mathematics
Used in Publishing Industry
Redes Inalámbricas
Ubicua/2006-2007
Redes
Why Use LaTeX?

Easy Publication
Easy Cross-References (table of contents, bibliography, figures,
tables, etc. --- no renumbering required!)
Clean, Tidy, Documents
Maintainable
Portable
Some conferences/journals require it!
Redes Inalámbricas
Ubicua/2006-2007
Redes
Getting Started with LaTeX

LaTeX Implementations
¾
¾

LaTeX Editors
¾
¾
¾

Your Favourite Text Editor
WinEdt http://www.winedt.com/
Emacs
WYSIWYG Font-ends for LaTex
¾
¾

LaTeX2e http://www.tug.org/
MiKTeX http://www.miktex.org/
LyX http://www.lyx.org/
TexMacs
Tools for Making Figures (eps)
¾
¾
¾
¾
Gnuplot
The Gimp
Microsoft Visio
OpenOffice
Redes Inalámbricas
Ubicua/2006-2007
Redes
Output Formats

LaTeX
¾
¾
¾

PDFLaTeX
¾

DVI
Postscript
PDF
PDF, Advanced Features
LaTeX2html
¾
Advanced HTML Output
Redes Inalámbricas
Ubicua/2006-2007
Redes
How To Write LaTeX Documents
Create a .tex file (e.g. thesis.tex)
Compile using latex

¾

View using xdvi
¾

xdvi thesis &
Or convert to postscript using dvips
¾

latex thesis.tex
dvips thesis.dvi > thesis.ps
Or convert to pdf using dvipdf
¾
dvipdf thesis.dvi
Redes Inalámbricas
Ubicua/2006-2007
Redes
Sample Proposal
\documentstyle[12pt]{report}
\usepackage[dvips]{epsfig}
\begin{document}
\input{epsf}
\title{Detection of Distributed Port Scans \\ A Thesis Proposal}
\author{Carrie Gates}
\date{10/2002}
\maketitle
\tableofcontents
\newpage
Redes Inalámbricas
Ubicua/2006-2007
Redes
Sample Proposal (cont’d)
%
%
%
%
\begin{abstract}
Brief description of my idea for detection of distributed port scans using
set covering techniques.
\end{abstract}
\input{introduction}
\input{related}
% reads in external file introduction.tex --- can be ispell’ed
% reads in external file related.tex
\appendix
\input appendix
% this will be labeled Appendix A.
% A second input file here would be labeled Appendix B.
\bibliographystyle{apalike}
\bibliography{proposal}
\end{document}
% also available: alpha, siam, plain
Redes Inalámbricas
Ubicua/2006-2007
Redes
introduction.tex
\chapter{Introduction}
\section{Problem Description}
\paragraph{}
The {\bf problem} to be addressed in this thesis is how can distributed port scans be
recognized as one collective port scan, as opposed to multiple independent port scans?
Info on port scans can be found in Section \ref{scans}.
\paragraph{}
Using Brinkley and Schell's classification scheme\cite{brinkley95a}, port scans are ``direct
probes’’, so are not in and of themselves dangerous. To quote Brinkley and
Schell\cite{brinkley95a}:
\begin{quote}
We must assume that a serious penetration attempt will be indirect in nature, will not
require direct physical access by the penetrator and/or operator to the penetrated target,
and will not advertise its presence or cause easily observable disturbances to the
system's behaviour….
\end{quote}
Redes Inalámbricas
Ubicua/2006-2007
Redes
related.tex
\chapter{Literature Review}
\paragraph{}
This chapter will provide background on both the people involved in computer attacks
and the tools they use.
\section{Port Scans}\label{scans}
\paragraph{}
There are four types of scans, which are defined by the target information to be gathered:
\begin{itemize}
\item vertical scans,
\item horizontal scans,
\item strobe scans, and
\item block scans.
\end{itemize}
A vertical scan refers to scanning all of the ports on a single system to determine what
services that one system is running.
Redes Inalámbricas
Ubicua/2006-2007
Redes
related.tex (cont’d)
\subsection{Stealth Port Scans}
\paragraph{}
Figure \ref{svm} shows a support vector machine.
\epsfysize=5cm
\begin{figure}[t]
\center{
\leavevmode
\epsfbox{svm.eps}
\caption{A data set consisting of two classes, represented here as solid circles, can be
separated using a hyperplane. The dashed line shows the convex hulls for the two sets.}
\label{svm}
}
\end{figure}
\paragraph{}
This is some verbatim text:
\begin{verbatim}
alert tcp $EXTERNAL_NET 10101 -> $HOME_NET any
alert tcp $EXTERNAL_NET 31790 -> $HOME_NET 31789
\end{verbatim}
Redes Inalámbricas
Ubicua/2006-2007
Redes
related.tex (cont’d)
\begin{table}[t]
\begin{center}
\begin{tabular}{|l|l|l|} \hline
anomaly & self-learning
& non time series \\
&
&
\\ \cline{3-3}
&
& time series
\\ \cline{2-3}
& programmed
& descriptive stat \\
&
&
\\
&
&
\\ \cline{3-3}
&
& default deny
\\ \hline \hline
signature & programmed
& state-modelling \\
&
&
\\ \cline{3-3}
&
& expert-system
\\ \cline{3-3}
&
& string-matching \\ \cline{3-3}
&
& simple rule-based \\ \hline \hline
signature-inspired & self-learning & automatic feature selection \\ \hline
\end{tabular}
\caption{Taxonomy of Intrusion Detection Systems, adapted from
Axelsson\cite{axelsson00a}} \label{axtax}
\end{center}
\end{table}
Redes Inalámbricas
Ubicua/2006-2007
Redes
Bibliographies

Many possible types:
¾
¾
¾
¾
¾
¾
¾
¾
¾

article
book
inbook
inproceedings
mastersthesis
phdthesis
techreport
unpublished
misc
http://www.ucalgary.ca/it/TeXclassfiles/AIX-52.html
Redes Inalámbricas
Ubicua/2006-2007
Redes
Sample Article
@article{cortes95a,
AUTHOR = "Corinna Cortes and Vladimir Vapnik",
YEAR = 1995,
TITLE = "Support-Vector Networks",
JOURNAL = "Machine Learning",
VOLUME = 20,
NUMBER = 3,
PAGES = {273 -- 297}
}
All articles are contained in proposal.bib
Need to bibtex it (bibtex proposal.bib)
Can have ALL readings in bibtex file … only those referenced will be used to generate the dvi
file

Redes Inalámbricas
Ubicua/2006-2007
Redes
So…

Files:
¾
¾
¾
¾
¾

Proposal.tex
Introduction.tex
Related.tex
Appendix
Proposal.bib
Commands:
¾
latex proposal
¾ bibtex proposal
¾ latex proposal
¾ latex proposal
-
generates proposal.aux
generates proposal.bbl
reads from proposal.bbl
fixes cross-references
Redes Inalámbricas
Ubicua/2006-2007
Redes
Redes Inalámbricas
Ubicua/2006-2007
Redes
Redes Inalámbricas
Ubicua/2006-2007
Redes
Redes Inalámbricas
Ubicua/2006-2007
Redes
Redes Inalámbricas
Ubicua/2006-2007
Redes
Environments

Text
Math
¾
¾
\[\sum_{i=1}^n i = \frac{n(n+1)}{2}\]
Inline:
My equation is $\alpha = \frac{n}{2} \times \xi$.
¾
¾
\begin{equation} … \end{equation}
http://www.agu.org/symbols.html
Redes Inalámbricas
Ubicua/2006-2007
Redes
\begin{equation} … \end{equation}
\begin{equation}
x = \left\{
\begin{array}{lr}
y & \mbox{if $y > 0$} \\
z+y & \mbox{otherwise}
\end{array}
\right.
\end{equation}
Redes Inalámbricas
Ubicua/2006-2007
Redes
Where to Get Help

LaTeX Documentation on Locutus
¾

/opt/teTeX/share/texmf/doc
Documentation Online
¾
¾
¾
http://www.tug.org/
http://www.mscs.dal.ca/~clyde/dalthesis/
http://www.google.ca
Redes Inalámbricas
Ubicua/2006-2007
Redes
Examples

Windows (WinEdt)
Linux (Emacs)
Both (Lyx)
Redes Inalámbricas
Ubicua/2006-2007
Redes
Lyx

Combina las facilidades de Word con la tecnología Latex
Es un front-end “casi” WYSIWYG para escribir en Latex
Evita tener que dominar Latex ya que, en sí, es un nuevo
lenguaje de programación
Redes Inalámbricas
Ubicua/2006-2007
Redes
Uso de gnuplot

Estándar de-facto para generación de gráficas
Las graficas se generan mediante instrucciones
Adecuado para representación de resultados en forma de lineas
Poco adecuado para gráficas de barras, sectoriales, etc.
Permite generar gráficos en formato eps (necesario para los
papers: latex)
Redes Inalámbricas
Ubicua/2006-2007
Redes
Uso de gnuplot

Ejemplo:
reset
set terminal x11
set xlabel 'Number of legacy 802.11 nodes (%)'
set ylabel 'Throughput per AC (Mbit/s)'
set key top right
set xtics 10
plot [0:100] [0.10:0.9] 'Hyb-def_br.txt' u 1:($2*512*8/60e6) title 'AC_VO (Voice)' w lp,\
'Hyb-def_br.txt' u 1:($4*512*8/60e6) title 'AC_VI (Video)' w lp,\
'Hyb-def_br.txt' u 1:($6*512*8/60e6) title 'AC_BE (Best effort)' w lp,\
'Hyb-def_br.txt' u 1:($8*512*8/60e6) title 'AC_BK (Background)' w lp
set terminal postscript eps
set out 'Hyb-def_br.eps'
replot
Redes Inalámbricas
Ubicua/2006-2007
Redes
Uso de gnuplot

Pagina web de Gnuplot: http://www.gnuplot.info/

Incluye:
¾
¾
¾
¾
¾
¾
¾
¾
¾
FAQ
Download
Demos and Screenshots
Documentation
Links
Useful scripts
Tutorials, learning, help
Building from CVS source
More on patching and building

Tema 6.- Publicación de resultados científicos

Transcripción

Documentos relacionados

SECURITY FOR UBIQUITOUS COMPUTING

organigrama de la división de electrónica y computación

CECILIA, Maximiliano Andrés - Universidad Nacional de Tucumán