Sources

The data for this project comes from the US Accidents(2016 - 2023) dataset available on Kaggle. This dataset includes detailed information on over 7 million traffic accidents across the United States, covering variables such as location, time, weather conditions, and accident severity.

To prepare the data for analysis, I selected a manageable subset of the full dataset to avoid memory issues and improve performance. Basic data cleaning was performed, including handling missing values, parsing dates and times, and selecting relevant variables for visualization and analysis. No major filtering or removal of observations was done beyond these standard preprocessing steps.