Lots of organizations and groups are interested in helping refugees. In fact, many apps and websites already exist to help refugees find things they need. Yet these resources all fail to address two key problems:
1) They are not addressing those who need help the most:
“Not all have access to this technology. It is no surprise that the refugees carrying smartphones are those who could afford them back home….And even those who have phones don’t always have access to the Internet, or to apps and services that might help them.” -- Wired, The Smart Phone Completely Changed the Refugee Crisis
2) The organizations do not have a way to receive input from refugees or coordinate their efforts with other groups
With these problems in mind, we designed a new framework to address these essential problem spaces.
What it does
We've created a website where refugees can SMS a phone number asking for help or information. For example, a refugee could ask: “What is the weather going to be like for the next week?” or, “Where can I find water?”.
Using machine learning, our system will identify keywords and automatically ask clarifying questions. Next, our website would automatically sort and convert these questions to "tickets" for organizations to fulfill. Organizations could send a reply like "We will be sending water to x location in 5 hours" or "You can find water at these 4 locations nearby".
If there was 50 tickets about water from the Moria refugee camp, that would indicate a large need that might indicate a direct intervention.
How I built it
We used the Twilio API to send and receive SMS. We also used Indico for text analysis and machine learning.
Challenges I ran into
Getting multiple API's and other tools to work together was a major challenge.
Accomplishments that I'm proud of
As a team, our proudest accomplishments are centered on creative collaboration and making safe spaces for learning. We hacked, made new friends, stacked cups, and had fun!
What I learned
Our team was comprised of members with diverse backgrounds. We're engineers, biologists, SLAM poets, people who literally learned to code over night, and most importantly people driven to create impact centered change through technology. We each went outside of our comfort zones and learned something completely new (using API's, Photoshop, Python, etc).
What's next for CoordinAid
In the future, we'd like to enable crowdsourcing through the public domain to better answer and address refugees' questions and needs. We'd also like to further machine learning so that question filtering could increase efficiency overall (no need to have humans answering every question).