ReLearn

Landing Page
Login Page
Account Settings Page
Tutor Matching Page

Motivation

Team BLNZ (creators of the ReLearn) were motivated to create a solution to the tutoring crisis we are currently undergoing. Despite the plethora of tutors and large student population, finding a compatible tutor-student match proved to be a challenge for many. Some basic challenges include the higher turnover and attrition rates that this industry face. As a result, students might end up getting overwhelmed and distressed over their studies. This is especially so in a pandemic-stricken world, where virtual learning has become the new normal. Our hack aims to solve all these problems for both educators and students by tackling the root cause of mismatched pairing.

What it does

ReLearn is a smooth and interactive experience for students to find the tutor(s) that best fit their needs for each subject category. Our database of tutor/student profiles is hosted on Google Cloud, and the secret to our special algorithm can be attributed to the use of machine learning models coupled with natural language processing! We provide a specially curated list of the best tutors available for each student through an intuitive UI. Additionally, updates can be made at any point of time as the application is backed by a persistent database.

How we built it

We split the tasks into two - namely frontend and backend, rather than a using a full-stack framework.

The Front End uses JavaScript.

It uses the framework React along side some smaller dependencies like: SASS, React-Notifications-Component, and EsLint. All these combine to provide a seamless user interface and comprehensive user feedback to improve the user experience.

The Back End uses Python.

We decided on Django and focused on processing the data needed to compare the tutors and students. We used machine learning methods and word processing algorithms, alongside personality descriptions to compare the profiles between the students and the tutors. A numerical score is generated through an appropriate weightage system, which aids in the curation of the tutor profiles that the student will see. As for storage purposes, we decided to leverage on Google Cloud due to its reliability and efficiency.

Challenges we ran into

Setting up the code and repositories on all our computers so that we all have the same environment took quite a bit of our time. Also, timezone issues were present as one of our members is based in Asia. Once we got that prepared, it was time to go to work! We made sure to get constant feedback on each new major commit that we did, this smoothed the development progress. It was also time-consuming to ensure that the entire architecture could reach the standard of seamless execution.

Accomplishments that we are proud of

Coming in this project, none of us were that confident. That being said, we are extremely satisfied with what we have accomplished being this project as a whole! As we created the project, every hole that we fell into was seen as a learning process and despite the multiple roadblocks, we managed to come up with a product that we are proud of. We found it satisfying to create something that has real world applications and actually solves multiple pain points in the academic world. Between the transitions on the front end , database handling/api calls on the backend and incorporating machine learning in our project (alongside the algorithms to compare tutors/students), we are unable to pinpoint exactly which area we are proud of because putting everything into motion is the biggest accomplishment for us all.

What we learned

This is the first time for most of us participating in a team and creating a product that we have to delegate and discuss among one-another. After doing this hackathon, we definitely learned how team work is important and how important it is to make comments to our commits to GitHub!!

Alongside the more social and team aspects of what we learned, we also learn a lot about Django as a whole (this was the first time for all of us using Django like this) and we learned about React as well. These two frameworks alongside the other dependencies were new territories to us and we learned a lot throughout the course of these short 36 hours.

What's next for ReLearn

Right now, ReLearn only has so much in terms of feedback. ReLearn accepts feedback of tutors in the form of reviews, these are only to be analyzed in our backend-not visible to the user. Making visible and comprehensive reports of the feedback of tutors would improve the quality of the experience for each student, so they can best pick the one that is right for them.These feedbacks would also be hosted on Google Cloud and then processed in our algorithms in our backend. Also, a dashboard system that has Google Analytics is in the works to ensure meaningful metrics for both educators and students - this will allow both groups to understand the needs and demands of the target audience.

Built With

Submitted to

Knight Hacks

Created by

I worked on planning the project as well as designing the logo

Peter Johnston
I worked on the front-end. I just used tons and tons of React and I can't think about all the stateless (which ironically have state) components I made

Zackary Santana
I worked on the backend for this particular project. It was my first time working with the Django REST framework alongside with google cloud SQL databases. It has been an eye opening experience personally as my team members and I worked together from ideation to implementation in less than 36 hours.

Wei Xuan -
Passionate about new and emerging technologies.
I worked on the backend and tutor matching algorithm. This was the first time I worked with APIs, Google Cloud, and SQL databases. This was a great learning experience and taught me a lot about communicating between different parts of a project and databases.

Nicholas Gray
AI/ML developer who is excited to learn and meet new people!