Final Project
The final project should be something related to the topics we discussed this term, and you should definitely plan
on discussing it with me as soon as you have an idea of what you might like to do.
There will be a Milestone report due at midnight the Wednesday after break (11/29).
This should be a PDF (or IPYNB) file uploaded to Gradescope with the following information:
- Names of all team members involved. You may have up to 3 people on a team.
- Brief description of the project: What problem are you going to try to solve, and
why it is important? How does it relate to either your interests or to what we did
in this class?
- At least four (4) resources (blog posts, book chapters, research papers), at least one of which
is a research paper published in a journal or conference proceedings.
- Project Plan: Restate the problem to be solved, specify the data sources you intend to use,
what, if any, data wrangling you will need to do,
the methods/algorithms you plan to apply, how you will run the project (e.g., on SCC), and the evaluation strategy to use to determine how successfully
you solved the problem. Be sure to state, if there is more than one team member, how you will divide
up the work.
The final project deadline is the last day of final exams: Thursday 12/21. You will need to upload to Gradescope a revised and updated version of your Milestone document to make it into a final project report
which is structured like a conference paper:
- Title of Project, and names of all team members.
- One paragraph abstract summarizing the problem, approach taken, and results obtained.
- Presentation of the results, including any useful illustrations, pseudo-code, etc. necessary to understand what you have done.
This should be something that can be read separately by someone who does not know much about the topic: restate the problem
definition and why it is interesting, then summarize the approach taken and how you proceeded (this would be an updated
plan, after the project is done), and finally a presentation of your results, including illustrations and figures as needed,
how you evaluated the results, and how successfully you think you accomplished what you set out to do. A final paragraph
should explore what you might do next if you were continuing to work on this topic.
- All code (but NOT huge data sets), or a link to a github holding your data.
- A statement jointly agreed upon by all team members on how much each person contributed, and what they did.
Project Ideas
I'll be posting ideas for projects in the next few weeks. But you could simply take a homework that you enjoyed, and expand it to a project as follows:
- Consider multiple data sources and "compare and contrast" the results obtained over these various sets. Assign to different team members and
collaborate on the "C&C" for the conclusion.
- Consider multiple approaches/algorithms for a problem, and compare and contrast the results obtained.