AE 04: Build better data (I)

Note

This application exercise is completed in class and submitted via a worksheet.

Discussion questions

1. Identify at least four ways to represent the arrival_date column that would potentially be useful for predicting avg_price_per_room.

2. Arrange these data preprocessing/feature engineering steps in the correct order for a KNN model:

   • Center and scale numeric predictors
   • Remove highly correlated predictors
   • Convert arrival_date to indicators for day of week, month, year
   • Remove arrival_date
   • Indicators for holidays
   • Remove zero-variance predictors
   • Convert categorical predictors to binary indicators