Inspiration
Saving One Life manually collected information on prospective adopters. This vetting process was tedious as the assistant director would personally find profile information from social media, apartment complexes, and driving records.
What it does
The application consists of a back end that generates a compatibility score based on application answers; flags questionable social media posts; and organizes information in MongoDB. The front end imports and exports csv files; allows administration and normal user access to specific data; login and registration; and applicant lists.
How we built it
The front end was built with Python, Flask, HTML, Bootstrap, and JavaScript/JQuery in the Visual Studios IDE. The back end was built with Python, Flask, and MongoDB. Various Python libraries were utilized in the project.
Challenges we ran into
Developing machine learning algorithm and creating parameters without real datasets from the nonprofit.
Accomplishments that we're proud of
Created a foundation for machine learning in a nonprofit context through the development of an easy-use database, algorithm framework, and password secured user interface.
What we learned
Learned MongoDB, the python flask framework, and python Twitter API. Furthered knowledge in docker for security and deployment. Created the foundation for a full stack website that automates the Saving One Life application processes.
What's next for Adoption Candidate Tracker
Receive previous application forms from Saving One Life to fully train scoring algorithm. Integrate solution with Saving One Life’s software. Implement JWT controls and site settings.
Built With
Python, Docker, MongoDB, Bootstrap, JQuery
Log in or sign up for Devpost to join the conversation.