In this email give me the NUMBERS (from the list below) of your top 6 project choices, in order... something like this:
7 2 11 18 5 9I will assume left to right is top pick to lesser picks, so in the above top pick is project 7 and 6th pick is project 9.
Titles link to client presentations, if available.
I am a professor of biology with thousands of digital audio recordings of songs recorded from wild songbirds. Bird song is complex, but I have software that renders its acoustical complexities as a bitmap, with x and y values respectively corresponding to time and pitch and the gray-scale value as loudness. With your expertise, I hope to have an automated routine that will extract the three bitmap values for each pixel (x, y, gray-scale value) in order to generate a set of values at several time points in a song. This will allow me to do my own statistical routine to categorize the song types objectively and quickly. Ultimately, I hope to determine what the birds are "saying" with their song.
We want to building adaptive software for physical systems (like robots, drones and autonomous cars). Software engineering has traditionally focused on building software systems for the business and finance domains. However, cars, robots, drones, etc., are increasingly becoming highly software-intensive and most of the innovation in these domains is today software-centric. Does traditional software engineering practices extend to these domains? How does software for these physical systems that move and “do” things differ from banking and business software? This small project will illustrate some of these differences. The algorithm at the heart of such software is a “control strategy” that needs to adapt in response to changes in the environment. We will aim towards building such a software running on a distributed system containing multiple Raspberry Pis and communicating via Bluetooth links.
CUUE is a digital jukebox that queues music from Spotify and Apple Music. The target audience is everyone - however primarily the target user or beachhead customer would-be college students. Cuue could be used in any situation where music is played with a group of people - house parties, pools, ubers, car rides, bars, coffee shops, restaurants, Benny's, and even in the Quad.
We need an Android app (phone or tablet); an Apple/iOS app (phone or tablet). Students will need to use Apple Music and Spotify APIs, both of them should be public.
https://www.figma.com/proto/6KgBwzUhWGmU0TZuTp2hka/CUUE-Prototype?node-id=2%3A2&scaling=min-zoomskillED is a web-based platform that provides access to postsecondary, college alternatives to high school and non-traditional students. High school counselors can use the tool to provide options to students for whom college is not the best fit. Trade, tech, vocational, and apprenticeship programs can use the platform to increase enrollment in students traditionally underrepresented at their institution. SkillED promotes equity in higher education through:
• Providing specific information needed to compare different programs, such as dates, costs, locations, and completion rates • Offering a comprehensive list of programs local to the user • Connecting students with direct links to programs applications and enrollment
https://prezi.com/r0qblqpkd-ym/skilled/
BigWords is a website in use around NC and the worlds to help readers work on vocabulary development. It began when a student and I built a prototype site (bigwords.cs.unc.edu) using what was (at the time) the latest greatest web technology. Fortunately (or unfortunately for me) the clients love the site and continue to use it, even thouugh the site was intended as an experiment to investitage some technology. It needs a rewrite using much more modern and robust web technology so that it can be supported into the future.
The advantage for the team of this project is it is already fully specified. They should make a site that works just like this one using current web tech.
Technologies to use are newer web tech... HTML + CSS + Javascript on the browser and Python + SQL on the server side.
Try it at bigwords.cs.unc.edu or bigwords2.cs.unc.eduThere is a cubic metric boatload of stuff to collect and generate in order to find, organize, and manage COMP 523 clients and projects. I need a web-based system that will collect, maintain, and present these various types of information in order to create and manage the COMP 523 class in a more timely and convenient manner.
Client and project information. We need to collect client information and information about project ideas. This segment should gnerate a web page with a summary of all project proposals with all relevant links embedded (word docs, PPTs, web sites, etc.). Clients will be indentified by name, organization/department, email address. This section will also provide clients with an explanation of the process of proposing, as well as an explanation of client obgligations once a project is selected. We want clients to fill out this information themselves via the web interface.
Team scheduling. Students will be able to select prefereces from the web description of project ideas. Teams will be formed to maximize (as much as practical) student matches. Unselected projects will remain in the database for use in following semesters.
Team management. This section allows the instructor to enter grades and evals for each milestone a team must reach. It will be viewable by individual team (protected somehow).
Progressive Web Apps are a newer technology for making phone apps. The basic idea is to build a Web app that adheres to a few basic interface requirements and provides a few basic services, and have that web app load itself onto your phone the first time to visit the URL. It provides an alternative to hosting apps at the Apple Store or Google Play.
This project will take one of the ideas presented here, and build a second version of it... one that is implemented as a PWA. The goal will be to explore and learn about PWAs... the pros and cons. The secondary goal will be to give the client an alternate version of the requested software.
https://en.wikipedia.org/wiki/Progressive_web_application
https://www.pwabuilder.com/
Tarheel Coin. Let's get rich. No not really. But let's learn the details of digital currency by making one. Like BitCoin, we envision Tarheel Coin (THCoin, or whatever the team wants to call it) to be a digital currency that functions in a similar manner. We will need to define and implement the following functions:
The purpose of this system will be to use as a technical demo for COMP 210 (410) and so we will build code for the various components (block chain, etc) rather than use a package designed to create digital currencies.
The system we implement will be a set of services, and we will need a web interface (at least) to access and use those services.
Over years of teaching COMP 410 (soon to be 210) I have accumulated a small heap (pun!) of example code for the various flavors of various data structures we study.
I need a web-based tool that will allow the gathering, showcasing, and exploration of these (and other) demo codes. I would like to be able to do several things:
Collaboration, communication, and related interpersonal skills are increasingly recognized as important domains of student competency as well as key attributes sought by employers. However, there currently does not exist a modular and easily portable tool for incorporating structured, small-group collaborations into online education. The current project seeks to fill role, so that educators can easily design and implement collaborative activities, and students can develop the skills required for effective teamwork.
This explains the Accessible Block Coding App
Thrive is a fully customizable app that provides structure, reinforces positive behaviors, promotes self-regulation, and offers safety reassurances all within one app. It is designed to assist children who are on the autism spectrum.
This will be an iPhone / iPad app.
City News Beat, Inc. owns and operates the streaming Television channels Tar Heel News Beat, Seattle News Beat, NYC News Beat and Bay Area News Beat for Roku users. Incubated as part of Cohort 11 in 2019 @Launch-Chapel Hill, a not-for-profit organization funded by the Town of Chapel Hill, Orange County and the University of North Carolina, City News Beat is committed to delivering an agenda free view of local news and weather for the cord-cutter and Smart TV communities they serve across the USA.
We create local video news stories and deliver them to Roku and Amazon Fire TV users who do not have cable TV. We have some basic software that allows us to load our content and schedule our content but we must rely on an outside vendor Zype.com to connect our content from AWS to our Roku Apps which we also built. The problem I want to solve is not just replacing what Zype does - I want A.I to help deliver a custom news experience for each user.
All of our users download our app because of the local news angle - something not well done on Roku yet by TV or Newspaper companies. We provide local stories, but then the individual preferences on content begin to creep into the user experience and we have trouble retaining people for longer than 16 minutes. We are ad supported so it is important to deliver meaningful content. Some users like to see food content recipes, others want to see events, others local sports, others pet content.
Each Roku and Fire TV has a unique Device ID. We would like to create an A.I. engine that can read different data sources including questions asked at the time the app is loaded, matches 3rd party data from a large reseller like Blue Kai to help us know more about the device users interests. We would like to connect that 3rd party interest data to our library, then deliver each user a Newscast-for-1 based on those preferences.
I believe that if we can create a custom experience our short form content becomes a compelling reason to stream City News Beat stories longer.
The UNC Physics and Astronomy department has created a database for exam questions and associated information, but an interface is needed for faculty to utilize this database effectively. We are looking for a team to create a web interface that can facilitate input and output with the database. In addition to the basic information access and review functions, client will develop different reports (exams) to be generated, as well as other functions such as showing questions my frequency of use, age in DB, etc.
It is well known that sedentary behavior (for example prolonged sitting) is a risk factor for cardiovascular and metabolic diseases such as stroke and type 2 diabetes. There is now overwhelming evidence that increasing physical activity levels, as well as decreasing sedentary behaviors, benefit physiological and psychological health. Furthermore, sitting for as little as 40 minutes has been shown to cause vascular dysfunction. The Exercise and Sport Science (EXSS) department is planning a pilot study to test the feasibility of, and adherence to, a work-based intermittent music program to decrease sedentary behavior amongst the EXSS administrative staff. Last year’s COMP 523 class began this work with the development of a computer-based music website. Essentially the project’s current stage is an itunes-like party shuffle type of program, in which a song only comes on at user-designated intervals as a friendly reminder to stand up/stretch/walk etc. Simply put, the program would be a party shuffle with gaps of silence. The other component of the study on the technological side of things is that we are planning on using songs that these staff members have requested (we’ve already collected quite a bit of their song requests). In this way, it would be a self-determined database of their favorite songs that will come on periodically throughout the day to hopefully decrease sedentary behavior, improve physical and mental wellbeing, and enhance their work environment. While last year’s class successfully developed the website, it appears the IP address has since expired, the website-user interface needs updating, the song database needs expansion, and links to scientific sources regarding the deleterious effects of sedentary behavior and prolonged sitting must be added. All of the technological components of project (e.g. passwords, coding information, etc) from last year are on hand and accessible. Lastly, this project has the potential to be written up into a short manuscript and/or be presented at a conference if the pilot work is successful.
I will focus on the need to provide an engaging training/game that encourages research/lab staff to follow through on EHS Biological Safety Officers recommendations concerning proper disposal of hazardous waste products.
All Facilities Services staff are required to complete annual Bloodborne Pathogen Safety Training. I have attached a copy of the PPT used for the Housekeeping Department's annual training. Slides 25 - 55 have the most pertinent information regarding the concerns for staff wellbeing when cleaning research labs.
There are several Skilled Trades Shops(e.g., plumbing, HVAC, electrical) and Housekeeping Zones responsible for providing maintenance and custodial services to these labs.
The failure of lab/research staff to dispose of waste products properly can expose other employees (particularly Housekeeping staff) to Bloodborne Pathogens (BBP). At this time Erika England, the Biological Safety Officer responsible for conducting this training has instructed Housekeepers to refrain from touching and emptying certain waste receptacles if they have not been disposed of and marked in the proper manner.
In this laboratory context, we are proposing a fun interaction, or maybe game, for training staff in proper procedures for safety. I will plan to cover the risks identified in the PPT and emphasize the need for additional but fun/engaging resources for the targeted population--i.e. research lab staff.
Jump Ahead is an after school program focused on building healthy communities through Jump Rope and interactive snack activities. We currently run our program in person at local elementary schools and are looking to expand to a digital platform. With this platform we aim to deliver our program to a broader population and serve as a revenue source for our in-person program.
Students and their parents need to go through middle men and locate tutors on their own. Students need the ability to locate highly rated tutors quickly and easily for help with academic subjects and standardized test prep. By connecting tutors with students academic achievement will increase along with financial independence for tutors and teachers. Current Solution: Currently we are tutors and managing the flow of students ourselves using calendly. We would love to have our own application that we can introduce other tutors into. Uber for Tutors perhaps.
Who are the users? The users of the software would be college student tutors looking to connect with students in the Raleigh Durham area. In addition students' parents may use the application to connect with tutors and schedule sessions for their children. Technology we envision includes Web browser (desktop); Android app (phone or tablet); Apple/iOS app (phone or tablet).
My research involves experimental work with zooplankton (think, sea monkeys) and we are interested in using species characteristics to predict impacts of climate change (extinction, range shifts, invasion). In practice, this means that we take many photos of individual zooplankton, and then make multiple morphological measurements using the pictures (currently using ImageJ).
Unfortunately this is a task that can involve thousands of images of different species, and currently requires training undergraduates to do by hand. It seems like an obvious target for automation! There are a number of issues that make this difficult, but that I think could be addressed: one is that images are supposed to show zooplankton that are flat on their sides, but some rotation is common that could probably be corrected for; the other is that there are multiple species and stages which might need to be accounted for in the software.
The upside is that there are 1000s of training images available, many of which have already been measured with a high level of confidence by hand. I had a student last year who built a simple ANN for classifying images into species, and that worked quite successfully, so the images are of good enough quality to support machine learning approaches, etc. (A few images can be seen on my website, carolinemtucker.com.). The resulting software would save hundreds of hours of labor and hopefully would be more accurate than relying on a large number of human users.
The system would be a web app and/or a Windows desktop app.
Examples: image-1 image-2