About

This is an example of using the Spark-ml library to solve the Titanic challenge on Kaggle.

I obtained a public score of 0.79426 on Kaggle with the current setup.

Using Zeppelin

You can start Zeppelin with docker-compose to explore the data with a titanic notebook already created.

docker-compose up -d

References:

• http://www.ippon.tech/training/big-data-with-spark/
• https://benfradet.github.io/blog/2015/12/16/Exploring-spark.ml-with-the-Titanic-Kaggle-competition
• https://www.kaggle.com/mrisdal/titanic/exploring-survival-on-the-titanic