The Erdős Institute, Data Science Boot Camp (Spring 2025)
Team: William Braverman, Dr. Justin Fong, Raul Hernandez-Gonzalez, Ayman Hussein
Mentor: Dr. Marcos Ortiz
Project Overview:
Cycling is a popular mode of transportation and recreation for many Americans. While it promotes outdoor activity, fitness, and environmental sustainability, it also comes with inherent risks. In particular, collisions involving bicycles and motor vehicles result in tens of thousands of injuries annually.
This project explores publicly available bicycle crash data from Chapel Hill, North Carolina, covering over 10,000 incidents between 2007 and 2018. By analyzing crash data in relation to cyclist demographics, road features, and weather conditions, we aim to identify risk factors and offer actionable insights to reduce crash severity and improve cycling safety.
Our primary objectives are twofold:
- To help cyclists choose safer routes based on road and weather conditions.
- To assist local transit authorities in making informed decisions about infrastructure design and improvements.
Project Goals:
- Route Safety Prediction: Identify safer biking routes under varying environmental and road conditions.
- Injury Severity Modeling: Develop a classification model to predict the likelihood of severe injuries and ambulance responses based on crash-related features.
Stakeholders:
- Cyclists and Navigation Apps (e.g., Strava): Can use the model to select safer routes for commuting or training.
- Local Transit Authorities: Can utilize the insights for strategic planning and enhancement of road safety infrastructure.