Inspiration
I am a transfer student who has always found it hard to discern which courses to take, with which professor, and what other students' experiences were. Just as I had gotten some of that finally figured out at my old college, I had started the transition to Mason where I knew I'd need to start that process again from scratch. Having to collect that data through word-of-mouth from fellow classmates felt impossible when it was hard enough to just make friends as a transfer commuter student.
There HAD to be some sort of solution to this problem that's keeping students blind and uneasy, resulting in higher failure-rates that are left unknown to others. All that's seen is another Mason student who's overwhelmed with a course that has an invisible failure-rate of over 40%, and them leaving the university due to too many failed or dropped classes.
I'm passionate about making education more accessible to fellow students, and increasing retention rates for universities - starting right here at Mason.
What it does
MasonMetric is a data-driven Academic Intelligence Suite that can convert complex, public university grade data into clear insights for both students and administrators. Students can use the interactive dashboard to visualize grade distributions by professor and course, empowering them with the ability to build balanced, successful schedules. For administrators, the platform serves as a Anomaly Detection Engine that can easily identify "bottleneck", or as we call them: "gatekeeper", courses with disproportionately high failure rates. By providing this level of transparency, we envision higher student success rates and assisting institutions improve their long-term retention rates in tandem.
How we built it
We developed the frontend using React, Vite, and Tailwind CSS to create a responsive and accessible UI that features dynamic pie charts and leaderboards. On the backend we used Python and the Pandas library to process raw CSV datasetss into clean, usable JSON files that could be easily read. These data structures were then for cloud storage that was scalable.
Challenges we ran into
One of the biggest challenges was the overall data engineering when it came to cleaning and normalizing large datasets to make sure they could be accurately visualized in real-time. I had to carefully consider the legal and ethical complexities of educational data. In order to be fully FERPA compliant, and still provide highly valuable insights, I focused solely on using de-identified, aggregate public data. I saw the vision of this project becoming a sellable "SaaS" startup model, so I had to rethink the whole UI/UX to accommodate both students and administrative needs down the line.
Accomplishments that we're proud of
I found the transitioning of our local static files to a live MongoDB Atlas cluster to be a huge major technical milestone for this project, too. I had the pleasure of using the MongoDB technology more with this project and I'm excited to use it again in the future.
What we learned
Developing this project taught me how important it is to be able to tell a story with data. Raw numbers are only valuable when they serve as solutions to specific problems, such as university retention rates. I delved deep into full-stack integration, specifically in connecting a React frontend with a NoSQL cloud backend. I also learned a plethora of new information on how I could pitch a hackathon project as a viable startup in the tech world.
What's next for MasonMetric
The first step is to create a standardized data pipeline that another institution could adopt with their own unique datasets. We want to eventually integrate with major Student Information Systems (SIS) to become a standard tool for universities can leverage.
Log in or sign up for Devpost to join the conversation.