This paper categorizes the U. S. states by frequency of fatal car accidents and distance traveled annually in order to analyze car accidents. The purpose is to improve the road safety situation through the analysis of patterns in data and provide useful recommendations to policy makers.
Our methodology involves:
- Data Collection: Compiling information on car crashes as well as annual miles from each state and sorting it into a data set format.
- Data Merging: Combining the datasets in order to obtain an idea on car accidents que from another angle.
- Feature Engineering: Creating other features, for example, the percentage of the driver that died in fatal accidents per thousand miles driven.
- Clustering Analysis: By clustering techniques to perform classification of the states on the basis of their accident history.
- Visualization: Employing big data analysis and trying to highlight the differences and similarities between distinct states.
The discovery shows that there are different clusters of states with drivers’ risky behaviours and car accident deaths rates being significantly different. The results derived from this research may well be of enormous relevance to policy makers in steering interventions and finances in a way that would promote road safety. Thus, the study demonstrates the importance of data analysis in addressing the important issues of public health and safety.