Fatal Collisions Prediction is a machine learning project aimed at classifying accidents as fatal or non-fatal using key accident-related features. The project utilizes data preprocessing, feature engineering, and model training with Logistic Regression, Decision Tree, and Support Vector Machine (SVM). The final model selection was based on hyperparameter tuning using Grid Search, with SVM emerging as the best-performing model. The pipeline includes SMOTE for handling class imbalance, and models are evaluated using precision, recall, and F1-score metrics to ensure accuracy and reliability in predictions.

This was a group project during my third semester, and as a result, I do not have sole rights to host it. However, you can access the source code on my Github Repository.