Skip to content

eliannemejia/OpenPort

 
 

Repository files navigation

OPENPORT

In a world increasingly torn by conflict, our proof of concept app OPENPORT is a digital lifeline for those navigating the asylum process. Whether you're a displaced refugee, an EU commissioner allocating aid, or a lawyer offering pro bono support, our platform delivers tailored insights from our extensive global asylum database.

Our app features 2 machine learning models, a logistic regression model for the asylum seeker persona, and a time series autoregression model under the EU commissioner persona. Both models used data from the Eurostat API. The only other page with real data is the "See Decision Stats per Capita" page under the EU Commissioner persona. It contains data from the Eurostat API and the World Bank. All other data shown on the app is mock data.

Our team

Mahika Modi, Alp Berrak, Jessica Poblete, Elianne Mejia

How to build the containers../

Before running the containers, make sure you create the following secret file in the secrets/ directory:

  1. .env
    • Contains environment variables for MySQL database setup
    • Example contents:
      MYSQL_ROOT_PASSWORD=password
      MYSQL_DATABASE=rethread_db
      MYSQL_USER=rethread_user
      MYSQL_PASSWORD=rethread_pass

Starting the Containers

To start the backend, database, and frontend:

docker compose up -d

Inside the root folder, create a .env file with the following contents: DB_USER=root DB_HOST=db DB_PORT=3306 DB_NAME=openport_database

Start Docker Containers Use Docker Compose to start all containers: docker compose up -d

Team Blog Posts

https://droidbait66.github.io/Team10/

Major Contributions

Jessica Poblete: Logistic regression model and UI for the app Elianne Mejia: Asylum persona and the connections between the lawyer persona and asylum persona Alp Berrak: Commissioner persona and the connections between the lawyer persona and comissioner persona Mahika Modi: Time series autoregression model and preprocessing of the logistic regression model's data

About

No description, website, or topics provided.

Resources

Stars

1 star

Watchers

0 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages

  • HTML 67.3%
  • Jupyter Notebook 30.2%
  • Python 2.5%