This is the code of my first submission to the Titanic Problem. It got a score of 0.8 and a rank of rougly 1000. I'm making it public. Upon improvement, i.e. dividing the dataset into pieces per title and tickets and applying random Forest to each segment the accuracy can be improved to 0.91(Within top 50 ranks). It's all a game of maintaining probability. This is my blog on the same https://utkarshraiblog.wordpress.com/2017/04/13/there-wasnt-enough-space-for-jack-but-could-you-have-survived-the-titanic/
utkarshrai/Titanic_Problem_Attempt1
Folders and files
| Name | Name | Last commit date | ||
|---|---|---|---|---|