This repo contains code and write-ups used to complete the capstone project for IBM's Data Science Professional Certificate. This exercise classifies accident severity based on various features pertaining to conditions around road and vehicles/traffic. A blog post detailing the effort and results can be seen here: [https://towardsdatascience.com/predicting-vehicle-accident-severity-using-ensemble-classifiers-and-automl-17ced9495932]
Defining the problem
Data visuaization using seaborn and matplotlib
Feature engineering and selection
Model fitting and training using ensemble classifiers, decision trees, and AutoML
Conclusions and recommendations for client