Final Project

Course Info

Instructor

Prof. Rosalind W. Picard

Departments

Media Arts and Sciences

As Taught In

Fall 2015

Level

Graduate

Topics

Learning Resource Types

Problem Sets

Lecture Notes

Written Assignments

Download Course

You will be required to complete a project for the course. Initial project ideas will be presented on Session 3. By Session 4, your project plan is due. Project progress presentations will occur on Session 7.

COUHES for Class Projects (for MIT students)

All projects that involve research on human participants must have the prior approval of the MIT Committee on the Use of Humans as Experimental Subjects (COUHES). MIT students were also required to pass the “CITI Certification”.

Project Examples

You will present the first half of your final project on Session 11, including:

What is the main problem you are addressing or the main idea you are exploring?
What is the closest background work that has been done?
What have you built / designed / examined? For example, if you have an experiment design, show us a picture of the design - perhaps a visual for how the study is set up. Or show us a demo video or live demo of the system you built (visuals / demos are strongly preferred to verbal descriptions).
What part of this is hardest? Illustrate clearly a challenge you have.

You will present the second half of your final project on Session 12, including:

(Brief recap) What is the main problem you are addressing or the main idea you are exploring? (Do not be afraid of Repetition, it is good for memory formation - set the context.)
How did you evaluate your work - What tests did you run? How many people? How were they recruited? What did they do? How do you know your test was good - what are your criteria for success?
What results did you find / learn / conclude? Show evidence / statistics to support your work.
Discuss your results: What is strong? What is weak? What would you recommend doing differently if you had time to do it over or if somebody else takes a pass at it? Help us learn from what you learned.

Write a page or two for your project proposal:

What is the goal of the project?
What parts (if any) of it exist now, and what parts need to be done to reach the goal? What is your back-up plan if any of these parts are at risk?
Who will do which parts? (if a group project).
Make it obvious what it has to do with Affective Computing. For example, you can use machine learning to learn about affect, and perhaps classify it, but the emphasis in this class should be not on a machine learning algorithm but instead on the way you are handling the affective data and its quirks / special nature. Similarly for human-computer interaction or education related projects, be sure to emphasize the affect-related issues.
Are you involving human subjects? Have you got COUHES approval or have you gotten your advisor to add you to an approved protocol, etc.? Often an existing protocol can be slightly modified, and this is fast. Or are you exempt? If in doubt, go to COUHES and ask.
Future (after the project is done): Visualize the big win—that you learn something you can teach the rest of us. It doesn’t have to be a big insight (although that’s awesome if it is) but it does have to be “done” in the sense that you have to get some answers. Did you build something and it does NOT work? That’s OK, let’s hear what you tried and why you think it didn’t work. If it does work, how do you know it works and when do you think it would break? What do you know now that you didn’t? You can’t answer this at this early stage of project planning (unless you’re cheating, but be a better person than that!). Let’s work together to design a project that will get some answers, so we can all learn from it.

Finally: Due Session 12, your final report (in PubPub or a .docx or .tex and .pdf of 4-7 pages)

Aim for no more than 4 pages if you’re a soloist and no more than 7 if you’re on a project team. If not using PubPub, then use a format for a conference of your choosing. For example, you could use the format for the Affective Computing and Intelligent Interaction Conference, which uses the IEEE Computer Society Formatting.