Accident Severity Prediction This repository contains a Python script for predicting the severity of accidents based on various traffic features. The prediction models implemented include Decision Tree, Logistic Regression, Support Vector Machine, Neural Network, Naive Bayes, Random Forest, Ensemble methods (Voting, Boosting, Bagging), and K-Nearest Neighbors.
Dependencies Google Colab Pandas Scikit-learn NumPy This dataset contains information about traffic accidents and their severity. The dataset should have the following columns:
Amenity, Bump, Crossing, Give_Way, Junction, No_Exit, Railway, Roundabout, Station, Stop, Traffic_Calming, Traffic_Signal, Turning_Loop, and Severity
Run the Python script in Google Colab. The script will read the dataset from your Google Drive, preprocess the data, and train various machine learning models for accident severity prediction. The performance of each model will be evaluated in terms of accuracy, precision, recall, and F1-score. Models Implemented The following models have been implemented and evaluated in the script:
Decision Tree Classifier Logistic Regression Support Vector Machine (SVM) Neural Network (Multilayer Perceptron) Naive Bayes (Gaussian) Random Forest Classifier Ensemble methods Voting (Logistic Regression, Random Forest, Gaussian Naive Bayes) Boosting (AdaBoost) Bagging (K-Neighbors Classifier) K-Nearest Neighbors (KNN) Results The script will output the accuracy, precision, recall, and F1-score for each model. You can compare these values to determine the best model for predicting accident severity based on the given features.