Real-Time 911 Prediction

Python
Database
Software Engineering
Full Stack
Machine Learning (ML)
Algorithms
HCI / UX / UI
API
Problem Solving
React
Spring 2025

A full-stack web application that provides forecasting of 911 EMS call volumes and locations. The backend is equipped with a PostgreSQL database and machine learning predictive engine. This engine leverages a time-series forecasting model(XGboost and Holt-Winters), specifically trained on historical call data, to identify temporal patterns and predict future EMS call volumes with greater accuracy. The backend then supports users to visualize demand hotspots on the frontend through an interactive heatmap. The application allows users to input a date range to generate and visualize expected call volumes across a City. With expected call volumes, users can make informed decisions on staffing and unit positioning.

1 Lifts 

Artifacts

Name Description
Demo of Final project for 911 prediction This video demonstrates our final project. Due to our NDA, we are unable to show the codebase, but the video covers our implementations, achievements, and key changes made throughout the project.   Link
Poster for the 911 Prediction Project This poster highlights the key achievements and milestones accomplished during the development of this project.   Download
Feedback