This project performs Exploratory Data Analysis (EDA) on a dataset of US road accidents to uncover patterns, trends, and insights related to accidents across different states, times, and environmental conditions.
The US Accident dataset contains data collected from various sources such as traffic cameras, weather sensors, and emergency calls. The goal is to understand how different factors contribute to accidents in the US using visual analysis.
- Analyzed over 2 million accident records
- Created interactive and static graphs showing:
- Top states with the most accidents
- Impact of weather, time of day, and day of the week
- Severity level distribution
- Accidents vs city population
- Monthly and yearly trends
- Accident hotspots on US map
- Python
- Pandas, NumPy
- Matplotlib, Seaborn, Plotly
- Jupyter Notebook
- Streamlit(dashboard)
- Source: US Accident Data (Kaggle)
- Format: CSV
- Size: ~2.8 million records
- 📍 Heatmaps of accident frequency by state
- 📅 Line plots of accidents per hour/day
- ☁️ Pie charts of accidents under weather conditions
- 🗺️ Maps showing accident clusters
- 📊 Bar charts for top cities with highest incidents
- California, Florida, and Texas report the most accidents.
- Most accidents occur during rush hours (7–9 AM and 4–6 PM).
- Weather conditions like fog and rain slightly increase accident risk.
- Majority of accidents are of severity level 2 (moderate).
- Fridays and Mondays show high accident counts.
- There are 60 insights in the notebook and aslo in dashboard
- github_name: aliahmad552
- linked: https://www.linkedin.com/in/ali-ahmad-dawana/
- Clone the repository
git clone https://github.com/aliahmad552/US-Accident-Report.git
cd US-Accident-Report