EZ Score

Inspiration

One thing I remember from my High School Statistics class was the look of exhaustion on my teachers face as he passed back our graded tests. He would tell the class just how long it would take him to grade these tests. It was the same process over and over and over again. Hearing this made me sad. What else could my teacher do. He has to test us on the material and he has to grade our tests so we can learn what we need to work on. The Artificial Intelligence boom had just started to take place during this time. A branch of Artificial Intelligence which I saw applicable to this problem was OCR and computer vision. I envisioned an app which allowed a teacher to feed in their grading criteria. Then they would simply upload a pdf of the student's work and then a Machine Learning model would grade the paper off a criteria and provide a score and feedback. The name EZ Score stems from the term "Z Score" in statistics. Since the inspiration for this project came to me in my Statistics class I thought why not name the product after it.

What it does

Currently our app allows a teacher to feed in their grading criteria into a text box. From there they would simply upload a pdf of the student's work and then a Machine Learning model would grade the paper off a criteria and provide a score and feedback. Currently our application is best suited for essays.

How we built it

We built EZ Score using Visual Studio code. After designing a rough plan of how the software was going to work, we split up the tasks into 3 categories. Frontend backend and connecting the two. From there the rest is history.

Challenges we ran into

One challenged we ran into involved connecting the front end and back end using Flask. At a few points throughout our project we struggled with having the backend effectively communicate with the frontend to display the frontend while having the functionality of the backend.

Accomplishments that we're proud of

We are proud that we were able to debug the code so it worked. When we saw that the frontend and backend weren't connecting properly we were discouraged but we hung in there and were able to get a working program.

What we learned

We learned a lot about working as a team and more importantly, creating a big application from scratch. In our classes so far, we never had to build an application from the ground up but we got our first taste of it here.

What's next for EZ Score

We're brimming with excitement for the promising future of EZ Score. Though our focus was on delivering a functional product within the tight timeframe of the hackathon, we see this as merely the beginning. Throughout the event, ideas for additional features flooded our minds, envisioning EZ Score evolving into a comprehensive tool ready for market penetration. Indeed, our conviction in the transformative power of Artificial Intelligence within education administration fuels our ambition. By leveraging AI to train models using past assessments, we foresee EZ Score becoming an indispensable asset. Our roadmap extends to enhancing optical character recognition, enabling grading of diverse subjects like math and science. As we refine our platform's interface and integrate a user authentication system, our ultimate aim crystallizes: to approach educational institutions with a compelling proposal for integrating our innovative app into their curriculum.

Built With

• css
• dropbox
• flask
• google-cloud
• html
• javascript
• openai
• openaiapi
• python