MindFULL+

Inspiration

The question of how to use technology to provide quality education for a population in need is complicated. We believe that technology can be used to enhance current educational systems to provide a better learning experience in the classroom. One major aspect that current educational systems lack is an effective way to obtain accurate and useful feedback from students regarding their education. Not every student excels in a typical learning environment, and not every student has found a learning method that works well for them. If teachers were able to get this necessary information (especially when it comes to student mental health), they could tailor their classes to help students learn better. This idea to use technology to improve classroom communication and feedback led to the creation of MindFULL+

What it does

MindFULL+ is an application intended to support communication between teachers, students, and parents by using AI sentiment analysis to evaluate student responses to polls regarding classes and teaching methods. Teachers can create classes (such as a Math or Social class) and post question polls for those classes, such as "How did you feel about the use of a whiteboard over a traditional chalkboard for this class". Students (and their parents) are able to see their enrolled classes and the polls posted by teachers for each class. They then enter in-text responses for these polls which are analyzed by FastAI to obtain a response of how each student really feels regarding each of their classes (and the teaching methods employed therein). This is done in the hopes that teachers can obtain valuable feedback to improve teaching methods, and class atmosphere and even help with budgeting for classroom supplies

For example:

Create a table of student responses and sentiment:

I liked today's math class because I got to use blocks to perform addition. The lecture was confusing, and hard to follow because I can't see the board. | Method | Feeling | Sentiment | Reason |

| --- | --- | --- | --- | | Blocks | Good | Positive | Addition | | Lecture | Confusing | Negative | Can't see board |

The app provides useful information to the teacher regarding the most well-received teaching methods. This can help the education system/teacher prioritize their budget. The app also provides insight into the individual level of the students, including which students are struggling and which are succeeding, and why.

How we built it

Our technologies included the following: JS, Python, React, Typescript, Postgresql, Styled Components, Docker, FastAPI, Express, openai ap, Prisma, vite.

We began the project first by brainstorming possible ideas and discussing the technologies we are familiar with. We created an ERD(entity relationship diagram) to understand the relationship between the different tables. We first broke into groups to handle different sections of the project, we later worked across the stack and spent a lot of time pair-programming.

Challenges we ran into

We ran into challenges regarding new technologies like FastAPI and issues with databases like setting up Docker with Postgress and technical difficulties with version control. With a name like Force Push Main who would have guessed we would have had git issues. Luckily they were resolved in the end but slowed our momentum.

Accomplishments that we're proud of

Completing the MVP Successfully used openai_api for the sentiment analysis. Learning a bunch of technologies in a short time. UI design and color scheme were inspired by the Japanese culture of simplicity and minimalism.

What we learned

We learned a lot about the potential of AI while experimenting with openai. We learned that we only spend less than 10% of our pre-adult life preparing for our future

What's next for MindFULL+

Given more time, we would like to take our app further by adding more features to increase parent user usability, enhanced profile customization, user anonymity, and the implementation of video responses with computer vision-assisted analysis.

Built With

• css
• express.js
• fastaiapi
• html
• javascript
• prisma
• python
• react
• sql
• styled-components
• typescript