Mini-Projects

CS 6451: HCC1
« class homepage

Suggested Mini-Projects at Bottom of Page

Due: Weeks 5, 10, and 15 (before Tuesday class)

Summary, Purpose, and Parameters
The purpose of the mini-project is to allow you to explore the breadth of research that is being performed by faculty in the Human-Centered Computing program. You may want to make it a goal of your mini-project to explore a discipline that is outside of your immediate research interest or area. This exploratory process serves several important functions:

  1. It gives you an excuse to introduce yourself to an HCC faculty member. You may find this relationship outside of your research area to be a useful independent source of advice throughout your graduate career.
  2. It introduces you to research outside your immediate area. This breadth of exposure may prove useful in your own research. For example, you may ultimately apply techniques you learn from your mini-project area to your own research area at some later point. Who knows: the mini-project may even become your dissertation topic!
  3. It allows faculty outside of your area to meet you. You can ultimately benefit from impressing a broader range of faculty (think career connections, etc.).
  4. It’s exploratory and fun! You will ultimately narrow down into a very specific research problem and area, and you won’t get too many chances like this to explore research in other areas.

It is up to you to:

  1. select a mini-project advisor, who can help you define the mini-project;
  2. define the mini-project;
  3. complete the mini-project to the satisfaction of the mini-project advisor (and me).

The specific milestones for the project, and specification of those milestones, is included below.

The scope of the project should be something that you can complete in 3–4 weeks as a “class project.” Make sure that you and your mini-project advisor are clear on these expectations.

You are required to perform three mini-projects. In terms of grading criteria below, your two highest grades will count (dropping the lowest). You are still required to perform all three. Remember, the purpose of these projects is to give you a “rotation” with a faculty member, who will turn out to be anyone from a colleague to potentially your PhD advisor.

Milestones and Important Dates
You can start your mini-project anytime, as long as long as they are completed by the due dates on the schedule.

You must provide a title for your mini-project and the name of the faculty member with whom you are working by the due date below.

Grading for this project will be A, B, or F. In addition to the initial report of who you are working with and the topic you have chosen, your final mini-project grade will have two components:

  1. We will ask the mini-project advisor for a grade (A or F) for the project and a brief evaluation/summary of your work. Be sure that the advisor name and email is in your project report! This component will form 75% of your mini-project grade. Your mini-project advisor will email this to the instructor.
  2. Your final mini-project report is a one-page summary of the mini-project: (1) the problem/topic, (2) what you did, and (3) what you learned from the project. This component of the project will also have an A or F component. This component will form 25% of your mini-project grade.

Mini-Project Suggestions

Literature Review: HCC Perspectives on Rural Networking

Mini-project: co-supervised by Grinter and Zegura

Background: Beki & Ellen have started a project (in collaboration with PSU and UCSB colleagues) focused on implementing a whitespaces network on a tribal reservation. Whitespaces uses frequencies allocated, but not used, for broadcasting services, allowing them to be used for other purposes. We plan to create a network using those frequencies that can be used in places where traditional Internet technologies are difficult to deploy—starting with a Native American Reservation.

The Mini-Project: We would like a student to review related HCC literature. What do we know about rural networking? Rural computing? What do people want? We recommend starting with these two papers below, both of which focus on the reservation we will be working with. You should read them, and follow the citation trails (what papers do they cite, who cites their papers—you can use Google Scholar and the ACM DL to get this information).

Sandvig, C. (2012). Connection at Ewiiaapaayp Mountain: Indigenous Internet Infrastructure. In: L. Nakamura & P. Chow-White (eds.) Race After the Internet. New York: Routledge.
http://www-personal.umich.edu/~csandvig/research/Sandvig__Connection_at_Ewiiaapaayp_Mountain.pdf

M. Vigil, E. Belding and M. Rantanen, “Repurposing FM: Radio Nowhere to OSNs Everywhere”, ACM CSCW, San Francisco, CA, February 2016
http://people.cs.ucsb.edu/ebelding/sites/people/ebelding/files/publications/cscw_radio.pdf

Deliverable: An annotated bibliography of papers that are relevant to understanding rural computing/networking. Literature reviews are a critical skill to develop as an HCC PhD student, so this project will provide you with an opportunity to develop this skill.

Helping Patients Manage Chronic Illness

Rosa I. Arriaga, Ph.D.

Both HCI researchers and Behavioral Scientist develop technology to help patients manage their chronic illness. In our lab we have conducted a systematic analyses of the literature on mobile apps for diabetes and have found some interesting parallels and disconnects between these two research communities. In this mini project you will be asked to update the literature review to include studies conducted over the past 3 years. You will also be asked to update our paper with the new data and prepare it for submission for a journal article. A high caliber mini project will lead to an authorship opportunity. 

Feedback on Online Behavior and SRS

Amy Bruckman

Does receiving feedback on online behavior actually change an individual’s future behavior? In this mini project, you will complete your human subjects training, and then interview three to five people whose postings have been featured on the site Shit Reddit Says (SRS), https://www.reddit.com/r/ShitRedditSays/ Do people even know their posting is linked on SRS? How do they feel about it? What can we learn from this more broadly about how to make the Internet a more civil place?

AI and Cognitive Science Projects with Ashok Goel

  A. Research Area: Artificial Intelligence and Machine Learning
 
A.1. Project name: Extracting Conceptual Knowledge for Engineering Design

Project description: Engineering design requires conceptual, causal and functional knowledge. This project seeks to automatically acquire this knowledge from natural language documents. It combines knowledge representation, analogical reasoning, natural language processing, and machine learning for the knowledge acquisition.   Contact names: Ashok Goel and Spencer Rugaber
Contact email: goel@cc.gatech.edu
Possible funding: No
Possible course credit: Yes

A.2. Project name: Exploiting Big Data for Scientific Discovery

Project description: The process of scientific discovery engages qualitative modeling as well as quantitative simulation, both of which require access to big data. This project seeks to use big data about biological species for qualitative modeling and quantitative simulation of ecological systems.

Contact names: Ashok Goel
Contact email: goel@cc.gatech.edu
Possible funding: No
Possible course credit: Yes

A.3. Project name: Explaining Intelligent Agents
Project description: Humans can explain their goals, knowledge, reasoning, decisions, actions – but computers cannot. But explanation is central to communication, trust, diagnosis, etc. So how can we build intelligent agents that can explain themselves?   Contact names: Ashok Goel
Contact email: goel@cc.gatech.edu
Possible funding: No
Possible course credit: Yes

A.4. Project name: Using Watson for Question Answering
Project description: This project builds on IBM’s Watson tool to build conversational systems that can use information of different types to answer questions.

Contact names: Ashok Goel
Contact email: goel@cc.gatech.edu
Possible funding: No
Possible course credit: Yes

B. Human-Centered Computing (Cognitive Science and Learning Science)

B.1. Project name: Learning about Scientific Discovery
Project description: How do scientists make discoveries? How can we help college-level biology students learn about the processes of scientific discovery? This project combines AI with cognitive science and learning science.

Contact names: Ashok Goel
Contact email: goel@cc.gatech.edu
Possible funding: No
Possible course credit: Yes

B.2. Project name: Perceptions of AI’s among Humans
Project description: How do students learn and interact with one another in online classes? How might we design virtual teaching assistants that can enhance these interactions?  How might the presence of AI agents impact human learning?

Contact names: Ashok Goel
Contact email: goel@cc.gatech.edu
Possible funding: No
Possible course credit: Yes