For df_calls, we will: Create new columns based on date, then join columns together to create columns such as SUB_YEAR_WEEK. We will create dummy variables for weekday names, months, and all call complaints. For df_disp, we will: We will merge df_disp_catso that we have consistent names for all dispositions. We will create dummy variables for all dispositions categories that have a frequency above 100 in the dataset