souptik4572 / HackOn-Submission-TraumaCheck

HackOn-Submission-TraumaCheck

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

TraumaCheck

Is your mental health is affected by your workplace environment? Let's find out...!

Inspiration

Due to this COVID-19 pandemic, we are currently noticing a large number of mental health breakdowns especially in people working from home. So, we decided to build a project that would let people know whether their mental health is interfering with their workplace performance.

What it does

This app takes some basic inputs from the user and analyses them through a pre-trained Machine Learning model to predict whether their mental health is interfering with their workplace performance.

How we built it

  • Gathered data regarding mental health from Kaggle
  • Trained various Machine Learning Models and calculated their accuracy
  • Choose our Random Forest Classifier model as the best one
  • Made an API server with FastAPI for processing data and wrapping our model
  • Built a Typeform-like form using React to gather user inputs

Challenges we ran into

  • Choosing a Machine Learning model with high enough accuracy
  • Building an API server for our model
  • Choosing memes that can make people smile :)
  • Tailoring React libraries for our needs

Accomplishments that we're proud of

We successfully overcame all the above-mentioned challenges and that's our biggest accomplishment! We also learnt how to train and deploy a Machine Learning model on a production server.

What we learned

  • A lot about python backend API microservices
  • Connecting a Machine Learning model with an API server for easy access from any sort of frontend framework
  • Reading open source code base and trying to modify/contribute

What's next for TraumaCheck

  • We'll try to suggest some solutions which will help people overcome mental health issues through the same app.
  • We'll try to implement an Unsupervised dynamic Machine Learning model in future.

About

HackOn-Submission-TraumaCheck


Languages

Language:JavaScript 97.7%Language:CSS 1.5%Language:HTML 0.8%