-
-
Recommender Site Page
Inspiration
Health profoundly impacts our lives, often enabling us to thrive but sometimes imposing restrictions. During this hackathon, we dedicated ourselves to developing a solution for those affected by medical conditions. Our goal was to create a tool that helps these individuals find the best places to live where they can continue to receive necessary treatment.
What it does
Our web tool collects relevant user data regarding their medical condition(s), income, and preferences to match them to cities across the United States.
How we built it
SettleSmart is developed as a Flask web application, leveraging MongoDB Atlas as the backbone for our backend operations. This robust database stores and manages our datasets along with user data, securely handled through Okta Auth0 for user authentication.
Our system is designed to interpret user data intelligently. By integrating straightforward but powerful models, SettleSmart assesses our comprehensive datasets to identify cities that align closely with our users’ needs. It evaluates key factors such as the availability of top-rated medical specialists and the cost of living, ensuring each recommendation is tailored to provide the best possible match for our users.
Challenges we ran into
Our greatest challenge was Dataset Research: One of our primary challenges was sourcing the data necessary for our models. Much of the data we needed was either scarce or expensive to access. This compelled us to creatively compile and feature-engineer a blend of alternative datasets that, while less direct, were crucial for the functionality of our application.
Accomplishments that we're proud of
We are most proud of our medical matching algorithms. While it is difficult to quantify what cities have the best treatment for a variety of medical condations, our utilization of a variety of datasources from repositories of medical practitioners, hospitals, and city costs created a similar quanitifier.
What we learned
Gained a better understand of the web frame stack we utilized during the project.
What's next for SettleSmart
- Include more datasets to better determine city preferences
- Build out a preference elicitation tool that utilizes user's pariwise selection to determine features that matter most to them.
Log in or sign up for Devpost to join the conversation.