Proposal

[dodgy719]

The project has a really great motivation! Couple of things:

• Be sure to note whether this is a predictive or inferential model (would recommend going with a predictive model)
• Addressing the class imbalance will be very important
• I will recommend using decision trees as a baseline and using a random forest model as they are more powerful and address a lot of the drawbacks of decision trees
• Lastly, you might want to look into creating bins for some of the categorical variables otherwise you might end up with a huge amount of columns

[mehince]

Thank you for the feedback - I will implement all of these changes into my proposal.