This is a Jupyter Notebook based on a public dataset. It was written sometime ago for my Nanodegree at Udacity.
Many other people conducted much better analysis (I believe the best one I saw was on Kaggle).
Anyway, maybe someone can find this interesting.