A preliminary exercice for an internship interview at Dataiku
My steps are described in the main.R file. To me, the most challenging part from this exercice was to understand the data, which was not always explicit. I worked a bit on improving the data but I think more work can be done on this dataset to improve my results. Especially concerning the ' ?' value. Finding a good way to select predictors was not simple either, the ones selected by the decision tree were a good beginning.
To put it in a nutshell, it was a very good exercice.
I hope my work will good enough to have an interview with a Dataiku member.