Rank 10 Public LB Solution for Hackathon:
- Created Date related features/ also Date difference features which shows how active the patient is
- Created Frequency related features
- Created category_category combine features
- Python for programming
- pandas and numpy libraries for methodology
- sklearn's logistic regression and lightgbm library for the model
- matplotlib and seaborn was used for plotting and analyzing the data
Seems like this hackathon is repeated after 4 years... and there are many solutions on them, I referred few.
Hackathon: Knocktober 2016
Rank: 10th on public LB and _ on private LB
LinktoLB