Accident Prone Area

Project Overview

The project talks about predicting traffic accidents. The output is categorized as accident severity on a different range, and the problem is handled as a machine learning classification problem. Bar graphs, line charts, histograms, and other visual aids have been used to illustrate the findings. Additionally, a map showing the locations of different accident-prone zones has been created. It was made possible by the dataset’s information on the latitude and longitude of these locations.

Tools & Technologies Used

Methodology / Process

Loading Datasets.
Data Preprocessing.
EDA and Visualizing Numerical Columns
Analysing Accident Frequency by weather conditions.
Analysing Accident Frequency by Hour of the Day.
Analysing Accident Frequency by Day of the Week.
Analysing High Accident Count and Severity.
Analysing of Accident Rates Urban and Rural Areas.
Analysing Accident Severity Distribution by Speed Limit.
Analysing Total Number of Casualties by Accident Severity
Creating and Comparing Machine learning models : Decision Tree , Random Forest, Logistic Regression, Ada Boost, LGBM, ANN.
Saving The Models and Analysing It To Find Users Prediction.