5.5.5.1 Unsupervised Historic Data Load

1. This is a pre-seeded batch and will be available in all workspaces (production & sandboxes)
2. This Batch is to be executed in the Sandbox workspace.

The historic data batch fetches 12 months or more of transactional data for all entities and will be aggregated for each entity. These aggregated features are used to populate the tables in the following with just one row for each entity.

The following tables that this batch will populate.

• AIF_BEHAVIORAL_DATA_UNSUP
• AIF_NON_BEHAVIORAL_DATA

These tables will be used for customer segmentation.

This batch has only one task defined under it:

• Historic_Data_Load

  Figure 5-80 Define Task

  
  

  
  Description of "Figure 5-80 Define Task"

  
  

Historic_Data_Load

• The objective folder for this task is

  Home/Modeling/Pipelines/AIF Batch Framework/Unsupervised ML/Historical
  Data.

• Do not change the parameters Objective, Model, Link types, and Synchronous Execution.
• The values in "Optional Parameters" can be edited:
  • model_group_name: Name of the model group the batch has to be run for as created in the admin notebook.
  • model_group_scenario_name: Name of the model group scenario under this model group for which the batch has to be run.
  • from_date: From date in DD-MON-YYYY format. Example: 01-Jul-2021
  • to_date: To date in DD-MON-YYYY format. Example: 31-Jul-2021
• Example: model_group_name=MODEL_GROUP_X,model_group_scenario_name=None,from_date =01-Jan-2020,to_date=31-Jan-2021