CISC 872 Semester Research Project
Spring 2005
Learning
Objectives: Graduate Students completing this project should
be able to independently and successfully:
(1) Perform a thorough bibliographical search
of a particular topic in computer science.
(2) Compare and contrast and summarize a set of
papers/research contributions on a particular topic.
(3) Write an informative, but concise and
critical survey of the state-of-the-art in a particular field of computer
science.
(4) Brainstorm new ideas on a particular topic,
based on previous work in the area.
(5) Develop an evaluation plan which would
successfully determine the effectiveness of new ideas.
(6) Write a well-developed, concise research
proposal.
(7) Learn to work together as a research team
in gathering information, discussing existing approaches to gain understanding,
critical thinking, and formulating new ideas.
(8) Critique a research proposal.
The
Divided Project: The following description lays out the various
subtasks and the timeline of minideadlines to ensure that each student makes
the best use of the entire semester to hit every aspect of the first stages of
research up through the proposal. Some
deliverables are done as a group effort while others are to be done
individually.
Potential
Topics: Based on current
proceedings and lectures in various conferences over the past year, I have
developed the following list of topics.
I prefer that you choose one of these topics, as they were included due
their “hot topic” status and the relevancy to this course. Only one group will focus on a given topic
from the list. If more than one group
selects the same topic, we will flip a coin.
This list of possible topics is given at <a
href=”proptopics.html”></a>.
If you are currently doing research in one
of these areas, and are already past the stage of creating a literature survey,
then you should choose another topic. If
you have just started research in one of these topics, but have not done a
literature survey, you may choose a topic somewhat closely related to your
topic or a completely unrelated topic.
If you are not sure which category you belong in here, you should talk
to me.
The rest of this document explains each deliverable in detail. For each deliverable, please submit a printed copy or email the pdf as an attachment. Please do not send the .dvi file.
Deadline 1 bibtex file first
draft: The goal for this
deadline is to identify the world for your topic; that is, search for all
papers relevant to your topic; not reading the papers, but determining
relevance based on title and abstract.
You should search in digital libraries (ACM, IEEE,...), recent
conferences and workshops that cover your topic, and then use the
bibliographies of recent papers to identify earlier relevant papers. The deliverable is a nicely formatted
reference list, created by using bibtex and latex, a paragraph that explains how you performed
your search, and one sentence describing the overall topic you are
investigating. This is a group effort,
with one deliverable for the whole group.
Deadline 2 Outline of
Literature Review: The
goal for this deliverable is to understand the timeline, overall contributions,
relative merits and limitations of the work embodied in the state-of-the-art in
your topic. You need to read only the
abstract, introduction, related work, and conclusions sections of each
paper. Do this reading in chronological
order (or reverse chronological order) of paper publication dates to obtain
some sense of how the research has evolved over the years. Then, develop an outline where you have
grouped the papers focusing on very similar problems, and then have a section
of the outline for each paper. For each
paper, be sure to include a subpart for problem addressed, contribution,
findings of any evaluation of the contribution, and limitations. Thus, your outline should look like:
This
outline should be in plain text so it is easy to insert into a latex file to
start writing. This is a group effort and has only one deliverable per group.
Deadline 3 First Draft of
background and literature survey:
The goal for this deadline is what would typically be found in the Background
and Related Work sections of a paper or proposal. You should read some Background and Related
Work sections of papers to see how they are written. That is, a good literature survey does not
just write a separate paragraph on every paper written in the field in any
order you want.
*
A good literature survey starts with a background section that familiarizes a
computer science reader to the basic topic area, such as testing web
applications – what constitutes a web application, how are they characterized,
examples from real life,...
*A
good literature survey tries to group papers addressing similar problems and
discuss and compare them together.
*A
good literature survey also presents the papers typically in some chronological
order within each problem identified.
*The
most relevant papers to what you want to focus on are presented first, and then
other papers that deal with problems related, but not so relevant, are
discussed very briefly and sometimes only cited as a group with a single
sentence. So, paragraphs are ordered
from most relevant work to least relevant work to your chosen problem of
interest.
*
A good literature survey will do the following for the most relevant papers:
describe
the overall goals/contributions of the paper, general approach and unique
characteristics of their approach, then end with restrictions/limitations of
that research. What didn’t they address?
Did they implement it and evaluate it?
A
related work section should be no more than 2 pages in the double column format
you are given in the latex file. Most
are more like 2 pages maximum. Each person in the group should write their
own background and related work sections.
Again, you can discuss what you think are the most relevant and least
relevant papers and agree on them as a group, but each member needs to learn
how to write these surveys by writing them on their own. It takes time and sometimes rewriting to put
it together as above. The deliverable is
a nicely formatted document with your Background section and your Related Work
section (done individually), and reference list. You should use latex for
formatting.
Deadline 4 Writeup of
brainstorming session: The
goal for this deadline is to get the group to:
* review and discuss the key
restrictions and limitations of existing research in the topic area as a whole
*
write a few paragraphs summarizing those limitations
*
discuss the key open issues/problems left unaddressed by existing work
*
write a few paragraphs summarizing those key open issues
*
brainstorm about possible approaches to attempt to address some subset of those
key issues: applying some technique used in solving other problems in other
domains, trying to develop new algorithms or alternate program
representations,...
*
write a few paragraphs outlining the possible approaches you could propose for
addressing these problems
*
think about how would you would evaluate the success of your proposed approach:
what are the appropriate questions to ask for evaluating the approach, what
could be implemented, what metrics could be measured and experiments could be
performed to judge whether your approach indeed addresses the problems or
improves on previous approaches in some way (space, execution time, program
analysis time,...)
The deliverable for this deadline
is a nicely formatted document that includes:
* your proposal title
* a section called Current
Limitations and Key Open Issues
* a section called Proposed
Research, which consists of two subsections:
* Proposed Novel
Approach(es)
* Evaluation Plans
This is a group deliverable.
You need only hand in one deliverable per group. These will be incorporated into each member’s
own individual proposal file for the next deadline.
Deadline 5 Complete Draft of
proposal: The goal for
this deadline is a complete first draft of a research proposal, created by
merging, integrating, and smoothing out your previous writings. This proposal should include the following
components:
* Proposal Title
* abstract (probably rewritten now
to reflect what you are addressing and your overall approach) (300 words or
less)
* introduction (motivate the general
topic and why it is worth studying, present overview of what goals your
research has) (1 page)
* background and state-of-the-art
(give the computer science reader some background on your general topic area,
as if they knew nothing about embedded systems or security,... and include your
related work section as part of this background now (getting rid of the related
work section title); end with a subsection called Limitations/unaddressed
problems if appropriate.) (2-3 pages).
So, this section includes your literature review (background and related
work) and your current limitations writeup from your brainstorming session. Be sure
to incorporate suggestions from the instructor on the background and related
work first draft.
* challenges and goals (enumerate
the challenges and goals that you are focusing on (1 page). this can include
the open issues that you wrote up for the brainstorming session.
* proposed research (present your
proposed approach(es), subsectioning as appropriate., including some steps you
would follow to do your research (1 page)
* evaluation plan (enumerate the
steps you will follow to do your
research)(no more than 1/2 page)
* summary of foreseen
contributions - how will your work help
society, summarize the contributions it would make if you indeed followed
through with this research (1 paragraph)
* reference list
This draft is written
individually by each member of the group, separately. The
deliverable is a nicely formatted first draft of a research proposal, one per
group member.
Deadline 6 Final version of
proposal: The goal for
this deadline is a rewritten, complete version of your research proposal,
focusing on the suggestions for improvement by me and your own observed
potential improvements. It has the same
format as the previous deadline, but now is single proposal
by the research group.
Each proposal will be reviewed by a set of students and discussed in
the research proposal panel session.
Please submit the final research proposal as a pdf file with the name of
your project as the name of the file, without any identification of the
researchers in the text or filename. Place the pdf file in the directory
called /usa/pollock/public/872pres/proposals.
Deadline 7 Proposal Reviews:
The goal for
this deadline is to learn how to review proposals, from the perspective of
a funding agency. Each person should sign up to review 3 proposals, making
sure we have at least 3 reviews per proposal. You should sign up by
editing the signup list in the proposal directory.
You should read the 3 proposals, fill out a review sheet, and bring it
to the panel review session ready to discuss the proposals.