B552: Knowledge-Based Computation
Spring 2008
Homework 3: Short Project Writeup
Due electronically as a pdf file before class on Wednesday, Oct 21.
Submission will be via oncourse.
This assignment involves writing a brief synopsis of your project
topic and plans. The purpose of the assignment is to help clarify the
goals and plans of your project and enable us to give useful
feedback. You're encouraged to discuss your plans and ideas with me
and other students (whether or not in your project group, or even
in the class!) before you submit your project plan.
Project groups should normally include one or two people. However, in
exceptional cases larger groups may be approved. Only one writeup
needs to be done per project group, and the writeup may be submitted
by any student in the group (please be sure to include the names of
all members in the group. The writeup should be approximately a
page or two long.
The Project
The goal of the project is to give you a chance to do research on a
topic of your choice---to critically examine an area, formulate
interesting questions, creatively try to overcome problems, and
analyze what you've learned from your work. In past classes, students
have sometimes chosen to continue these projects beyond the class, and
have even had their project work lead to research publications. Try
to select a challenging topic that you care about, and don't be afraid
to take risks!
Each project must include a substantial implementation of a
knowledge-based AI method, normally building on one of the
methods covered in class. It is acceptable to use existing code as
part of the project, if properly acknowledged. However, use of that
code does not decrease the amount of code to be developed by the
project team: a substantial
the project work must include a major software component and
evaluation.
There are both rich knowledge resources (e.g., OpenCyc, ConceptNet)
and software resources (e.g., Montylingua, the AI microprograms,
IUCBRF) available for use in your projects if you wish.
The Writeup
The writeup should begin with a project title concisely summarizing
your project. The writeup itself should describe your goals and
intended methods, should make clear what you are
investigating/modeling, what AI methods you intend to use, the
intended I/O behavior of the system, and the problems you would like
the work to address, and how you will evaluate the project's success.
It should also mention both why the problem is important, what you
expect to be the hardest part, and what you expect to demonstrate or
learn from the project.
Please be prepared to briefly describe your project to the class on
October 21.
Grading criteria for the project
To help in your project design, please keep in mind that the following
grading criteria will be used at the
end of the semester for the projects themselves. If you're interested
in a project that doesn't fit into all the aspects of this list,
please talk to me in advance.
- Problem (25%)
- Is it interesting and challenging?
- Model and assessment (35%)
- Is the theoretical solution interesting?
- Would it scale up?
- Does it make theoretical claims?
(E.g., saying something about needed knowledge for a task, or
the strengths and weaknesses of a given process, or about what
are the hard and easy parts of the problem you're attacking.)
Is its performance well evaluated?
- Paper (20%)
- Writing
- Motivation
- Relation to other work
- Analysis of strengths+weaknesses
- Clarity on program (All major points should be described, but at a
fairly high level)
- Program (20%)
- A well-documented implementation of model.
Please let me know if you have any questions!