EquiLearn

Inspiration

As women of color, we recognize the inequities and learning disparities within our education system. With our project, we hope to be the bridge that leads historically marginalized and disenfranchised students to an academically rich tomorrow.

What it does

EquiLearn is a web program that creates personalized study material for students based on their notes.

How we built it

We wanted to ensure user satisfaction with our innovation. On the front-end, using Figma to develop our prototype, languages/frameworks such as React, JavaScript, Tailwind CSS, and HTML all worked in the development.

To connect the front end to the back end, we used VS Code in Python to develop accessors and mutators that collects user input and sends it to our AI study guide maker to personalize the study guide. Parsing methods varied based on the input leading us to use Beautiful Soup and the requests library.

Creating the study guides, we utilized Hugging Face and the Transformers Library in our development of our AI study guide builder. Through disciplines of natural language processing, we used the tasks of text generation and summarization for our function.

Challenges we ran into

The largest challenge we encountered was developing our AI function with the Transformers Library as well as downloading the Transformers Library.

Accomplishments that we're proud of

We are proud of our commitment to our intention behind the project: decolonizing education. With this being the fuel to our drive, we engaged with new technology and software such as the Transformers AI Library and Beautiful Soup. Our endurance and our dedication to our innovation is what we are deeply proud of.

What we learned

We recognized the difficulty associated with engaging with new, and rather foreign, software. Downloading and processing through the terminal was new to all of us. Utilizing Bash in the command line was also a new skill we acquired through this project. On the front end, fiddled around with getting the local host to run. Ultimately, with our programming and computation, we learn new debugging skills we were otherwise never exposed to.

Another thing we learned is to persevere and take breaks. This project is something we all deeply cared about and wanted it to accomplish what it set out to do. We are proud with our comradery and our resilience through this process and staying true to our commitment of creating equitable learning.

What's next for EquiLearn

We hope to expand EquiLearn's ability. Considering the time, we were limited to specific file types, disciplines, and user input. In the future, we hope to enhance EquiLearn's abilities to take in greater file types, serve as a resource for all disciplines, fill in gaps within user input, and diversify output to varying learning styles.