• We can see people seeking financial or help of any kind inside the campus in real-time.
  • According to Wilder Research only Seven-County Metropolitan Area has 6763 homeless people.
  • This was especially evident in a population not in a formal shelter, which was up by 93%.

What it does

  • We introduce an incentive-based system to help the needy by helping users reach out to nearby homeless people.
  • Helping locate and contributing gives the users points which they can redeem later on to compete and claim prizes.

How we built it

  • The front end has been built in react-native using expo framework.
  • Backend is on Loopback in NodeJS using MongoDB for database needs.
  • We have built a TensorFlow image classifier in order to validate people who need help.
  • The TensorFlow API has been deployed using Flask.
  • We have used Heroku for deploying API endpoints.

Challenges we ran into

  • Scraping training data for image classifier for difficult to get.
  • Integrating MongoDB with loopback was also challenging.

What's next for Minnesota-Nice!

  • Focusing more on security concerns.
  • Providing better solutions for validation of contributions using machine learning.

Built With

Share this project: