Tag: Kaggle Competition

Using a shotgun approach for the Titanic competition

After completing Alexis Cook’s very useful Titanic Tutorial, I couldn’t help myself and spent a couple of days hacking around to try and improve my score without going through the usual data science workflow of EDA, feature engineering, model selection, hyperparameter tuning, train/test iterations.

I know it’s not the proper way of doing data science, but like I said, I just couldn’t help myself.

Continue reading