🚑 NYC EMS 911 Calls Analysis & Prediction

This project analyzes New York City EMS 911 call data to uncover patterns and build predictive models for emergency response.

🌐 Live Demo

👉 Open Streamlit App

📊 Project Goals

• Analyze call patterns by borough and time
• Identify trends and high-demand periods
• Predict future call volume using time series models
• Build machine learning models for response insights

🧠 Methods & Models

• Exploratory Data Analysis (EDA)
• SARIMA (time series forecasting)
• Random Forest (prediction)
• Feature Engineering

🛠️ Tech Stack

• Python (pandas, numpy, scikit-learn, statsmodels)
• SQL
• Tableau (dashboard)
• Streamlit (interactive app)

📁 Dataset

NYC EMS 911 Calls dataset (cleaned and processed)

📈 Key Outputs

• Borough-based call analysis
• Hourly and seasonal trends
• Forecasted EMS demand
• Interactive dashboard (Tableau / Streamlit)

📈 Key Insights

• EMS call volume is highest in Brooklyn compared to other boroughs
• Peak call hours occur during late afternoon and evening periods
• Seasonal patterns show increased demand during specific months
• SARIMA model successfully captures time-based trends in call volume
• Random Forest highlights key features influencing response patterns