Course Project
The project can take several forms:
- Solve an open theory problem, formulate a new problem, or make some other
contribution to the study of algorithms/lower bounds for processing big
data.
- Implement an algorithm and use it to analyze some massive dataset, or do a
comparison between implementations of various algorithms that solve the
same problem.
- Write a survey on a few related papers. A good approach is to first try to
solve an open problem, which generally requires reading several background
papers first, then switching to a survey if the problem evades solution.
You are encouraged to relate the final project to your research interests, and
you will not be limited to the topics discussed in class. “Big data” should
somehow be involved though.
Project Proposals:
You must submit a project proposal by Thursday, October 26, 2017. When
submitting project proposals, use the normal mechanism for submitting problem
sets. The proposal should be at most two pages long with 10pt font and 1 inch
margins.
We decide whether to approve your project’s theme based on the proposal, so it
is imperative that you do some serious thinking about the project before
writing the proposal. Even though you should not change the topic of the
project after submitting a proposal, you may change the form of the project
(such as writing a survey if you fail in bringing a theoretical contribution).
Policies:
- You may work in teams of up to three on the project. Each team only needs
to submit one project wite-up, and the presentation time will be split
evenly between the team members. Both team members receive the same grade
on the write-up portion of the project. You are allowed and encouraged to
consult with anybody, including the teaching staff, when working on your
project.
- You must submit a paper by 11am on December 8, 2017, which should be at
most 10 pages, excluding title page (which can include team member names,
title, and abstract) and bibliography. Use at least 10pt font and 1 inch
margins all around. A clearly marked appendix of unbounded length can
follow the bibliography, but there is no guarantee that any of it will be
read. As such, keep the most interesting parts of your work in the 10 page
body.
- As with problem sets, your paper must be typeset in LaTeX. Submit your
paper via email to
sketchingbigdata-f17-assignments@seas.harvard.edu with
filename of the form [first-initial][last-name]-project.pdf (e.g.
flast-project.pdf for First Last). If there are multiple members
on the team, only one submission need be made, and name the file using any
of the team members’ names (make sure all members’ names are written in the
paper though!).
- You are also required to do a presentation during the last couple lectures.