Displaying the head of each file  reveals a snapshot of the original 16 column names as well as a preview of the first 5 rows for each set. We drop the columns that are not of interest in order to make the computation more efficient keeping only the usertype column and date column.