The Austin Cyclist Safety project was initiated to address the growing concerns within the cycling community regarding safety on Austin's roads. Utilizing data on cyclist-related incidents, this analysis aims to provide insights into the factors contributing to these events and offer a predictive tool to assess risk levels for cyclists.
- Data Analysis: Conducted comprehensive analysis of cyclist incident data, focusing on factors such as time of day, location, and environmental conditions.
- Model Development: Implemented a Random Forest classifier enhanced with SMOTE (Synthetic Minority Over-sampling Technique) to address class imbalance, achieving an accuracy score indicative of the model's reliability in predicting incident severity.
- Risk Prediction Application: Developed an interactive Streamlit application that enables users to assess the risk of cycling based on specific conditions, offering valuable insights for safer cycling practices.
The culmination of this project is an accessible and user-friendly web application that provides cyclists with real-time risk assessments. By inputting factors such as the day of the week, time of day, and current weather conditions, users can receive a prediction on the risk level of cycling in specific conditions. This tool not only raises awareness about the safety of Austin's cycling environment but also empowers cyclists with the knowledge to make informed decisions about their travel.
We can access the application on the streamlit web by visiting the following link: https://bike-severity-application-fgpeaitrtcfy5y4tnrlbst.streamlit.app/