This page gives information on the course project for CS793.
For details on the paper critiques see here
Back to course homepage
Overview
- There are a number of different possibilities for a project in CS793. Here, I outline a few, but the students are encouraged to propose their own type of project.
- Projects will be presented by students to the class during the last week.
- Projects are worth 50\% of the final mark.
- To be done individually or in small groups (2-3 people). Group projects are expected to be proportionally more involved, but are encouraged.
Options
- Option A (Literature survey):
- Pick a problem in Health Informatics that interests you
- Search the literature for approaches to tackle this problem
- Survey and discuss the relative strengths of each approach
- Your survey will be evaluated based on its completeness (it covers all related work), and on the originality and usefulness of how it structures the relevant literature.
- Option B (Empirical evaluation):
- Pick a HI question that interests you, and for which solutions exist.
- Pick a solution method (already published or implemented)
- Study different methodologies for evaluating the method, and propose one that fits your question
- Evaluate the solution using your experimental methodology
- If your design or testing involves people (end-users), seek appropriate ethical approval from the Office of Research Ethics (ORE).
- Option C (Implementation):
- Choose an application area of health informatics.
- Conduct a literature review and propose a new application.
- Implement a prototype of your application and test it.
- Your prototype need not be fully functional (so can be a mock-up that is tested using e.g. a Wizard-of-Oz methodology).
Proposal (10% of final mark)
Deadline: Jan 31
In 1-2 pages, explain what your project will be about, and what methods you plan to use. Cite about 10 papers that you plan to survey (about 15 if you are doing a literature review).
Report (50% of final mark)
Deadline: April 11
The report should be about 20 pages long, and should be structured like a research paper.