Background: Many Americans choose cycling to commute to work or for recreation. It's a fun way to get outside, get some fresh air, and get exercise. But, all of this comes with some danger as bike crashes with motor vehicles account for tens of thousands of injuries per year. The most lethal year on record for bicycle crashes was 2022, with over 1000 cyclists losing their lives.

Goal:
- Provide data driven insights that cyclists can use to keep themselves safe on the road and that municipalities can use when designing infrastructure to make it more bike friendly.

Data:
- Clean and analyze bicycle crash data. This will be provided from Kaggle municipal DOTs. (Example: https://www.kaggle.com/datasets/adityadesai13/11000-bike-crash-data?resource=download)
- Environmental: how do time of day, day of week, weather, light, road conditions, etc. correlate with bike crash severity and frequency?
- Demographic: do things like age, sex, race, etc. play a role in the severity of injury?
- Infrastructure: what types of road and bicycle infrastructure keeps cyclists safe (or conversely what puts them at the most risk)?

Method:
- Use Linear Regression and Machine Learning models to determine what factors play the most significant role in cyclist safety.
- Train predictive model to estimate injury risk for cyclists based on their cycling conditions, route, etc.