Ideas for group projects
This page contains project ideas submitted by a variety of people. If you contribute an idea, please indicate who the client / contact is.
Also, check out the section-specific project idea page(s):
More recent ideas
Portable Online Identities (Added on Aug 25, 2006)
Doesn't it vex you when you try to access an interesting looking website and you realize that it requires you to fill out a registration form ("Free! Only 5 minutes!") in order to enter? Wouldn't it be nice if your computer or Web browser could remember that your name is Jane Doe and that you were born on January, 1st 1970 so that you didn't have to type it in at every site you came to? This problem is at the core of this project idea: design a way for people to store their online identities so that these people can easily join online communities.
Two big problems we might face are 1) Making these online identitites poortable among different computers, and 2) Making it hard for others to steal an identity.
My name is Oge Nnadi and I'm a member of this class. Please contact me at onnadi3@mail.gatech.edu if you want to discuss this idea. If you know you want to join, just add your name below
Members
Oge Nnadi
Jin Ah
Computer Aided Human-Human Interaction (Added on Aug 24 2006)
Motivation
1. I believe the IP communication may get far beyond instant messaging, chats or voice/video calling
2. There are some haptic devices [1] used in human-machine interaction, but, even though the ideas exist, I have not discovered any real piece of hardware used for interaction between people.
Idea
Do some initial reseach and tests in the area mentioned in the title. Create a device that lets you "feel" or "touch" the person on the other side of the wire by force feedback and/or electrical pulses. We can start with an existing device such as force-feedback mouse, if there are some hardware/El.Engineering people willing to participate it would be great to have our own device manufactured.
How it fits within the CS/PSYC6750
I don't know yet, open to discussion. Feasible options are likely (A) making an existing device work the way we want it and see how the users feel about it or (B) design and test a really simple prototype of our own.
Contact
Martin Bednar, MS HCI-CS student
bednar@gatech.edu
suggestions/opinions welcome
[1] http://haptic.mech.northwestern.edu/intro/gallery/
Mobile Office Applications (Added on 22nd Aug 2006)
1. Controlling presentation slides from mobile phone.
2. Calendar application which will send reminders of appointments to mobile phone.
3. Real-time Lecture feedback monitoring application.
4. Alerting user on Desktop PC about the incoming call on cell phone.
If interested in any of the above ideas or any idea on top of these ideas or anyother mobile application related idea shoot back with an email to mrunalk@gmail.com. I am MS-HCI student having 3 years experience in Graphics, Broadcasting and Newsroom automation products. I have experience in Windows SDK programming, OpenGL, Graphics card DDK programming, C++, MFC, VC++, Unix Shell scripting, Oracle-SQL,etc. I am looking for people with background in Mobile application programming experience in Symbian, WindowsCE or PocketPC. If you are interested any other Mobile application related ideas you are most welcome.
Experiment Observation Recording
In a darkened room we observe fish in a tank interact with each other. We are trying to solve the mystery of how they arrive at their "pecking order". We want a better computer program than the one we have now to record our observations in real time, make editions and additions later, and display the data. Examples of typical data entries: 1B2 (1 bites 2); 4 D 1 (4 displays to 1). The main problem with our current program is that it requires too much training for the many lab assistants who work for only a short time. It is slow, hard and boring for them to learn to quickly and accurately touch type the codes for the different animal behaviors. The program must run on a PC, and should use standard equipment or at least inexpensive equipment. This is our core problem. Not only don't we see any way to modify our current program to accomplish this, we don't know how to do this, period.
There are several other data entry features we need the program to have. Our current program essentially has these. It has the ability to go in and out of a correction mode easily as you enter data in real time, and a comment mode you can toggle in and out of quickly. I'm not including all of these in this writeup. I don't think these are very tough to handle. The point I want to make, because I think it will make the project more attractive to work on, is that we know pretty well what functions we want, at what level of speed or ease, because we've been collecting these data for many years and we have a program that we've been using for some time.
Our current program does not have the ability to go back to a data file and reenter more data. This would be very helpful to us.
We've invented a new way to display the data, inspired partly by musical notation. When we draw our data by hand with this notation, we can see what is happening dynamically in the group! We want the program to be able to use this notation in its displays. If the program would store data in a simple format, then we believe that many other researchers will use the program to display their own data. We also want some other kinds of data display.
Many other researchers collect and analyze data on group interactions. Subjects include birds, crayfish, fish, primates, and humans. In some cases, the program without change would be useful in other research; in other cases the program would have to handle some situations we don't encounter, such as groups larger than 4 or 5, or behaviors that have a "subject" but no "object" (modifier), e.g. "3 quivers". We've identified the key features like this which would enable the program to have wider impact, if possible. However, we'd be thrilled just to have our core problem solved!
We also have a list of bells and whistles which would be nice, but are not high priority. The top priority, again, is ease of training for accurate real time data collection in a dark room where the user is looking at the animals, not the screen or keyboard or joystick or microphone or whatever. The other functions, which our current program does not have, and which we particularly want, are the ability to display data with our music notation, and the ability to make additions or changes to an existing data file.
Here is an example of a whistle: the user has to be able to quickly flag a data entry as incorrect if she doesn't have time to fix it then because there is a flurry of activity. The program has to display the data collected so far during the session. A nice whistle would be if these functions were integrated, so that if "2 chases 1" at 0:14:20 has been flagged as incorrect, then in the display on the screen, then that datum is flagged in some way (colored? marked?). It would be even better if the user could, during a lull, point and click on (or otherwise specify) a displayed flagged datum and easily correct it.
Contact: Craig Tovey ctovey@isye.gatech.edu or tovey@cc.gatech.edu
x43034 Room 420 ISyE building
Additonal Contact: Ivan Chase ichase@notes.cc.sunysb.edu
Laboratory for Study of Social Organization
Dept. of Sociology SUNY Stony Brook
Searching, Organizing, and Browsing Data
Here's the basic idea. Mark Devaney (former PhD student) and I have
developed a J2EE-based engine which can search, cluster, sort, and analyze large amounts of data. It runs on a server and the client basically sends it requests and gets back results through a defined API.
For an end-user to use it, the client needs to run some kind of GUI through which the end-user can search the data (like doing a Google search, if the data is web documents, but it might be other stuff) and also access the other functionalities such as clustering and sorting. The project would be to build this GUI.
Separately, a GUI is also needed to build the databases that the engine runs on, if the user wants to run it on data other than the pre-analyzed data that is already provided. This may be part of the project or it may be a different project.
The engine is running at Enkia and we could provide APIs to a team who wanted to use it. I guess it would be somewhat like taking the raw Google APIs and building a whole new GUI for it, except that the engine provides more than just web search and so there are more complex workflow issues to deal with.
Client: Ashwin Ram (ashwin@cc.gatech.edu)
Enhancing Web Site Navigation to Improve User Retention and Service Utilization within a Multilingual Environment
MiDieta is a bilingual online health resource for consumers, implemented across multiple portals. Consumers have access to various services, including an articles database, an exercise and dietary evaluation, as well as a customized weight management program that supplies full menus. Services are tiered, so that some are available to the general public while others are available for a nominal fee.
In order to retain users, the site must not only contain satisfactory content, but the user must be able to easily access this content and incorporate it into their daily living. Navigation is critical to bring the user to the service they are looking for. If a user cannot navigate to their desired service, they may leave the site completely. If users do not advance to higher tiers of service, then the web service fails to generate income.
The goal of this project is to analyze user interaction with the web site and the user interface itself, and how it differs based on language & culture. Prototype and study various methods for enhancing the site navigation with the goal of increasing utilization of available services, conversion rates to higher tiers of service, and improving customer retention/renewal by increasing their perceived value of the resource.
http://christus.mydiet.md
Contact:
Lisa Wells
Sales & Marketing Manager
DrTango, Inc. - Hispanic Health
900 Old Roswell Lakes Parkway
Suite 230
Roswell, GA 30076
770-649-0298 x.108
770-649-0299 fax
www.drtango.com
Desktop + PDA Interfaces
Users of personal digital assistants (PDAs) like Pilots or PocketPCs will often dock their PDA when working on their desktop machine and then ignore it. We could instead take advantage of the PDA as an additional input/output device to monitor or control some aspect of the desktop computer. The goal of this project is to prototype and study several applications whose interfaces relies on adding a PDA to a desktop PC.
For example:
- Use the PDA to monitor your instant messaging contact list: who's online, who's busy, etc.
- Use the PDA to keep track of the progress of different tasks (ftping files, downloading web pages, print a paper).
- Use the PDA to switch between different projects (sending email, writing a paper, working on classwork).
- Use the PDA to monitor your hard disk usage, CPU activity, and network activity.
Client: Jeff Pierce (jpierce@cc.gatech.edu)
Augmenting PDAs with Displays in the Environment
A major drawback to personal digital assistants (PDAs) is their small screen size. While a PDA's computational power, memory, and battery life will all improve with time, the size of its screen is limited by what people are willing to carry with them. If we wish to avoid having the small screen size dictate the types of tasks we can use a PDA for, we need to find a way to improve the available screen real estate.
One way to add more screen real estate is to recognize that there are many places in the world around us where there are displays in the environment. For example, offices often contain desktop displays and classrooms often contain projectors. If we could temporarily annex a display in the environment when using our PDA, we could use our PDA for a wider range of tasks (or complete existing tasks more efficiently).
In addition to creating an infrastructure for annexing displays, we need to determine how to effectively use that additional screen real estate, for example by dividing application functionality across the different displays. The goal of this project is to prototype and study several applications that divide input and output to the application across (at least) a PDA screen and a desktop display.
Client: Jeff Pierce (jpierce@cc.gatech.edu)
Interfaces for Situation-Aware Audio Communication
We rely on intercom systems and telephones to speak with our associate in the neighboring office or a family member in another city. However, many phone calls do not reach the intended person or it may not be an opportune time for the recipient to talk. There are many technologies providing the location of a person or a personal device, such as PDA or phone. What would the interaction to create a remote conversation look like if it were not tied to a telephone number pad or a desktop monitor? How can location and other situation information be incorporated within the user interaction to aide in creating a distributed conversation? One example of an alternate interaction, is the augmentation of the
Digital Family Portrait with audio services (Mynatt and Rowan, Georgia Tech). A family member's photo is displayed on a touchscreen within a frame, serving as a decorative item for the home or office. When the image is touched, location information is used to determine whether or not to create an audio connection between the location of the digital portrait and the family member it displays. The image is replaced by another interface to portray activity of the called person, status of the connection, and controls for the audio connection. What interaction mechanisms would fit into our home, our car, our mobile lifestyle and provide useful information for initiating a distributed conversation?
Client: Kris Nagel (kris@cc.gatech.edu)
Improving Equation Editor Interfaces
Currently there are great ways to write high quality math equations and formulas, i.e. use Tex/Latex. However, the current input language for the Tex type systems are terrible. They use lots of complex \xxxxx type expressions that are hard to learn and remember. Also they make it very difficult for support staff to help with papers since Tex is like a foreign language to them.
Current Solutions:
1. Learn Tex.
2. Some smart editors do support symbol tables so with a mouse click one gets a
"framework" then then gets to be filled in.
Drawbacks:
1. Obvious
2. The mouse click is not easy. Lots of symbols and tables. Still need to know what is going on. Also from HCI point of view seems like one could argue that requires lots of human clicks and decisions and fine motor moves to make a simple formulae. Also one is forced to use their particular editor.
New Solution:
I have an idea that we could make a system that would be very smart and would greatly simplify the input of math. Its would still probably output Tex etc. so compatible with rest of the world. But would be easier to learn and use.
My idea is not completely worked out but is based on some "hand experiments" that I have done. Roughly the idea is to build a very simple interface that allows you to input the math symbols. The idea is this:
Suppose one is typing and now wants to type in a math symbol/expression. The user has a small number of what I call magic keys (perhaps ONE will work or perhaps as many as THREE but not tons of tables ...). By press one of the magic keys the system GUESSES what the previous typed symbol should change into. For example:
press "a magic" might create the greek symbol alpha.
The cool point is that the system will have a grammar for math that is nontrivial. The magic keys will move the user from one string to another within this grammar. Note, a VERY trivial version is already used in cell phones. Press abc key twice and get b.
Note, one of the cool parts of this project will be to figure out a "grammar" type system for math. It seems surprising to me that all the current methods do not seem to use any semantic info about math to cut down on human input. Clearly, the fact that math expressions are not random could tremendously cut down on the bits a human needs to input.
Client: Dick Lipton (rjl@cc.gatech.edu)
General Concept: Usable Wearable Computer
- Currently, wearable and portable computing (i.e. Palm-sized devices as opposed to notebook PCs) are both fast becoming crippled fields - the former because there are few wearable devices that can work without making the user look like an extra in a SF movie, the latter because their functionality is being rapidly absorbed by converged devices (which are less powerful in general but offer more desirable functions).
- Proposal is to prototype a device which could function as a straightforward 'wearable' computer possessing computing power approximately on the level of a Pentium II - possibly providing visual feedback through a highly simplified HMD (research must be done into the ergonomic and physical properties of this technology). The main concern is the design of the interface and especially of the input devices to be used to achieve maximum usability in all situations.
- Many other interesting functions could be achieved by such a technology - project should explore at least some of these and design suitable interfaces. For example, imagine you're in a foreign city and can't read a shop sign, and no one around speaks English. No problem: instead of fumbling for a dictionary, look at the sign and activate the OCR/Translate option (We need a better interface for this!). Your computer uses the built-in miniature camera (mobile phone-style, so it's not visible at all) on the HMD to take a picture of the sign, runs it through OCR, and translates it - within seconds, you have a much better idea of what the sign says.
Look up Nathanael Ng at LCC (nng@steel.lcc.gatech.edu) if interested.
Older ideas
We will be adding different ideas for possible group projects here. Anyone should feel free to post ideas here.
Where the heck is the bus!! (Bus tracking system). Design a way to track the buses on the GT bus service, and then display their routes, locations, and time to arrival at various places on campus (i.e., at the stops, at key buildings, etc.). Users would be all those who take the bus, and would interact with the system potentially at the bus stop, via the Web, over the phone, via a PDA, etc. Lots of angle to this.(Bruce Walker)
A similar system has been implemented in various European cities using SMS messaging systems (Helsinki, Bologna). The bus stop has a number. You send a message to the service center, with the bus stop no. and bus line you're waiting for, and the systems answers with an SMS telling you how many minutes the wait is, and other information. This systems can be actually implemented with really very little programming, just by using a cell phone connected to an infrared equipped PC. All kinds of other fancy stuff could be added: notification system, reservations, etc. (Giovanni Iachello)
- This system has already been implemented for GT's campus. Check out NextBus. However, I have a somewhat similar idea which I will post shortly. (Kyle Woodlock)
Server load sonification over the phone. Server administrators need to know what is going on with their servers, even when they are at home or traveling. However, they might not have Internet access where they are. How could they call in to a system, and hear the server load, review activity over the last few minutes, hours, days, etc.? (Bruce)
Stock data sonification over the phone See above, but the user is a day-trader who wants to know about the activities of his stock(s), but does not have access to the net (no visual interface). (Bruce)
Experimental email reader interface I have
a completed version of Ishmail, my absurdly personalized email reader for people getting hundreds of messages a day. It is a UI nightmare: rule maintainence, dozens of mailboxes, mailbox-level alarms, "personas" for different mailboxes, mailbox identity merging, logs of activity, archiving, search, mail deferral, managing outgoing as well as incoming mail, automatic deletion, short cuts and keybindings, etc, etc.
If nothing else, it really is a ubiquitous problem that no one seems to have gotten quite right yet. I've made some decisions and built an interface that mostly works for me, but to say its designed for the power user who needs features and who thinks emacs is the last word in WYSIWYG would be an understatement. I want someone to figure this out for me :). Two page report about project.
Client: Prof. Charles Isbell (isbell@cc) — Posted by Stasko
Interactive hall kiosk for CERCS We have what sounds like the perfect environment for synergy with your course project. The "video kiosk" outside of CCB207 is currently displaying the GVU "What's Happening" screensaver. This is meant to be a stop-gap measure until we get enough interactive content available. We have recently acquired a license for NetKey Creator (a simply Windows based
kiosk development environment centered on web content) and need to present an engaging interactive interface to current CERCS research and events. While we are good at the systems research stuff, GUI development (attract loops, content presentation, etc.) are not in our mainstream.
Client: Matthew Wolenetz, wolenetz@cc.gatech.edu — Posted by Stasko
Several Ideas: (still thinking of others)
- digital camera where you can send pics to friends as soon as u take them or post to the web, you can also do some photo editing on it....the camera can also receive photos from your friends...
- parking spot finder that can tell you where the best parking spot is and lets you reserve it depending on your proximity to the spot...that way u wont miss the previews of the movie or end up driving around endlessly in mall parking lots
- something with mobile phones like being able to buy things from a vending machine by beaming infrared from ur phone
- record store kiosk that allows you to browse music tracks, order concert tickets for your fav band, order cd's (if the store is out of stock), etc.
E-mail me if interested or have other cool ideas: Jehan Moghazy - jehan@cc.gatech.edu (Section B)
Interface for Surveillance cameras to catch illegal dumping in municipalities (Daniel Muller - Section B)
Having worked for the City of Atlanta, I know there is a terrible and costly problem caused by the illegal dumping of trash, debris, construction materials, tires, etc. in different problem areas around the city. My proposal is to develop an interface that would harvest the input from multiple surveillance cameras and direct this information to the Atlanta Police Department and/or Sanitation Department for prosecution. The surveillance feeds could be monitored for real time apprehension of violators (ie. immediate police dispatch) or video could be captured for use in tracking down and apprehending perpetrators after the fact and in court as evidence. You can e-mail me at gtg338j@mail.gatech.edu. Looking for team members with programming and CS background, since my own background is humanities related.
Reachablity manager Personal communication devices are becoming more and more popular, and with them rises the problem of being continously interrupted at unappropriate times. New, more powerful communication devices could allow the user to define his or her own reachability policies. E.g. not being disturbed by friends' calls during classes, meetings, etc. Redirecting work calls between 8pm and 8am. Blocking calls without caller ID, etc. Various implementations of such an idea exist, but are not very successful. Given the very complex policies and requirements, the design of such a system should be carefully planned, to avoid falling into the trap of developing something more difficult to use than just evaluating on a call by call basis whether to answer or not. This system could be designed and a partial implementation could be arranged for using simple software (e.g. java). Giovanni Iachello (email)
Room scheduling system So many new buildings on campus, no one knows what rooms are available for meetings, nor how (who) to reserve them. And what facilities does each one have (size, technology, etc)? How about a system that lets you see the rooms available on campus, perhaps graphically, maybe text format as well, sorted by ammenities and distance from your office; then you can put a room on "hold" somehow, as a message is sent to the coordinator for that room (usually a secretary in the building), who will then "confirm" your reservation, and a recipt will be sent back to you. Reminders could also be built in, as could a list of participants to notify (and remind!)....Could start with one bulding, then grow the system to include more of the campus...(Bruce)
Zoo Navigator I took my son to Zoo Atlanta many times. Of cource, zoo is a fun place for kids, but we couldn't fully enjoy all of the zoo.
Problems we encounterd were...
1. Sometimes we couldn't find any animals in their habitat.
2. Zookeepers feed gorillas once a day. But we never saw them because we didn't know the schedule and were not there on time.
Possible solutions
1. This was because animals might be just behind bush or something. If there were cameras to capture all scenes of the habitat, visitors could observe animals. And possibly, some kind of sensor or camera with computer vision techniques detects their exact locations. By the information, visitors can avoid wasting time by visiting empty exhibitions. Moreover information of each individual animal (eg. name, gender, story etc.) could be attached to the location data.
2. There are several demostrations and events at the zoo. Although the zoo provides such information via the web, such information should be available anywhere in the zoo.
Hand-held computer might be one of the best choices for this system. Rui Yamada(Section B)
A few ideas...
- using a GPS enabled cell phone, PDA, or laptop to track friends and other locations (who also have gps enabled devices).
- a cell phone adaptation of oscarweb or CS classes website (to access syllabi and things) etc...
- checkbook feature on a cell phone that could allow us to keep track of all the checkcard purchases we make on the go... could maybe interface with financial agencies to keep official balance current.
Some of these may require simulation as the tools to bring them to fruition may not be available. Nevertheless, we can still design a usable interface for them. (If any one was interested in doing the first idea on the page, Keeping track of the bus system, by Dr. Walker then let me know). Lamar Gardere from section B.
Website for a political talk show Goals of the site would include:
1.) Streaming audio and video content from the show.
2.) links to sites, information, and streaming media related to show content.
3.) A weblog so people can challenge, debate, refine, expand and dig deeper into issues raised on the show by contributing comments, links, articles, and audio and video.
An important goal of the site would be for the weblog aspect to significantly shape the content of the tv show by providing information, ideas, resources, contact information for potential guests, etc.
Note: I am currently producing a political talk show for the Atlanta public access cable channel. I would like to eventually start implementing such a web site for the show. I am producing the show with both formats (television and eventual web delivery) in mind. I would like to incorporate ideas from this class into the show this semester.
James Ransone
Salvation from Ludites
I'm not quite sure exactly what the project might be, but there is a need for bridging the technology gap between a person or business with all the leading edge technology and the low tech Ludites whith whom they must communicate/interface. For example, a web enabled business still needs to interface with customers who still use rotary dial phones on a party line. In a very short time it will become a major PITA to deal with the "legacy" systems and Neandertals in terms of productivity etc. If we could figure out how to interface hi-tech to lo-tech without slowing down or going into a completely different system, it would be really great. Rick Gordon Section A regordon@bellsouth.net or richard.gordon@gtri.gatech.edu
Mobile Situational Vizualization
We are looking for 2 more group members. We want to design an interface for some mobile situational vizualization application. One idea is a project that I am working on for my RA. We a building a system that will allow users to view information about other users and objects on campus. One scenario is a war game where users can track the locations of their friends and/or enemies, objects that they can pick up (weapons, health etc.), and campus data. Another possibility was the idea of tracking the GT buses. Open to other suggestions. Group members: Tazama St. Julien, stjulien@cc.gatech.edu & Lamar Gardere, gtg509k@mail.gatech.edu (Sect B). Need an HCI student.
Firefighter Training Virtual Environment
A virtual environment that simulates a single family home that is on fire. The user instructs teams of virtual firefighters to perform tasks to put out the fires. Currently the tasks are issued by typing commands. A gui prototype was developed, but was never actually incorporated. A better GUI for issuing commands to firefighters and monitoring the simulation could be a possible project. Possible for allready formed groups. Contact Tazama St. Julien, stjulien@cc.gatech.edu
Hi , I am Dr. Meghna Krishnan . I am interested in the topic of interface design for the elderly. I am a HCI(LCC) and I am also doing GRA in wireless technology for the people with disabilities. Elderly people are handicapped by their disabilities. I have some ideas like:-
1) A pill reminder for the elderly- elderly often end up with a situation where they take about 10-12 pills a day and that really is difficult for them to remember . How about a system that is handy to carry, easy to operate( keeping in mind their low vision ), voice interactive system that reminds them to take the pill, and that probably can also be a cellphone/PDA .... There are some devices in the market that are pill reminders but their interface can be improved a thousand times!!
2) another group of users could be the blind. Either a cell phone or a PDA or a pill reminder which comes with voice interaction and Easy access features. Please mail me at meghnakrishnan@hotmail.com