5.2.4.1 Behavioral Model Aggregate Base Features

  • This pre-seeded batch will be available in all the workspaces (Production and Sandboxes ).

    Note:

    This batch has to be executed in the Sandbox workspace.
  • This batch creates base features for scenario model training in the sandbox workspace.
Batch and Task Parameters
The batch contains a single task named Aggregate_Base_Features.

Figure 5-23 Define Task for Aggregate_Base_Features



Task: Aggregate_Base_Features, Task Parameters
  • Objective folder for this task:

    Home / Modeling / Pipelines / ML4AML / Scenario Model / Batch / Base Features

    Note:

    Do not change any parameter except Optional Parameters.
  • Optional Parameters:
    • model_group_name: Name of the Model Group for which Base Feature Aggregation is to be created. Example: LOB1.
    • model_name: Name of the Model used while importing the model template using Admin Notebook. Example: RMF.
    • from_date: The start date for the Historic Data lookup is in DD-MM-YYYY format.
    • to_date: End Date for Historic Data lookup in DD-MM-YYYY format.
    • prod_flag: Flag to indicate Training/Scoring scenario. The option is Y or N.
    • For sandbox/historic training scenarios, the prod_flag should be set to N.
    • include_full_lookback: Flag to indicate whether the lookback should consider data beyond the from_date to aggregating base features. The option is Y or N.
    • last_run_date: The last run date within the from_date and to_date range, which exactly matches the scenario run date in DD-MM-YYYY format.
    • frequency: The frequency of the scenario execution.

      For example: 1 (Daily), 7 (Weekly), 14 (Bi-weekly), 30/31 (Monthly).

    • look_back: The lookback period for the scenario. For example: 30.
    • focus: The model entity name is provided in the Admin notebook dataframe while creating the model group. The option is CUSTOMER or ACCOUNT.

      Figure 5-24 Parameters for Aggregate Base Features



      filters: Scenario specific parameters that are used to give additional control for the base feature aggregation. The format to be provided is as follows:

      Param1 : Value1 ~ Param2 : Value2a | Value2b | Value2c

      For example: PRIMARY_CUST_FL : Y ~ MANTAS_BUSINESS_ACCT_TYPES : RBK | RBR ~ INCL_CASH_TRXN_PRDCT_TYPE_LST:DEBIT-CARD|SVC|CREDIT-CARD|CURRENCY|PHYS

      Figure 5-25 Edit Task for Aggregate Base Features



Table 5-4 Task Parameters for Scenario Model Aggregate Base Features

Parameter Description
PRIMARY_CUST_FL It indicates what accounts are included by customer focus. The values are:
  • Y: Cover only accounts for which a customer plays a primary role.
  • N: Cover accounts over which a customer has discretion.
INCLUDE_B28_TRNFR_FL It controls the inclusion or exclusion of bank-to-bank transactions. The values are:
  • Y: Includes transactions with a bank-to-bank transfer.
  • N: Excludes transactions with a bank-to-bank transfer, and the originator or beneficiary is the ultimate originator or beneficiary of the funds (i.e., Pass Through Indicator is set to No).
INCLUDE_TRUSTED_TRANS_FL

It controls the inclusion or exclusion of transactions designated as trusted transactions.

Trusted transactions are those considered trusted based upon the presence of one or more trusted pairs (parties identified as enjoying a trusted relationship) on the transaction. The values are:
  • Y: Include trusted transactions.
  • N: Exclude trusted transactions.
INCL_RLTD_PARTIES It allows coverage of all transactions between related parties. The values are:
  • Y: Covers all transactions.
  • N: Excludes transactions between related parties.
RPTNG_CURR_FL The value is Y or N.

If Y, then all aggregation is to be done on reporting currency.

MIN_HRG_RISK_LVL Minimum list risk level greater than or equal to (>=) a transaction considered high risk.
INCL_SEC_PARTY_FL It controls the inclusion or exclusion of secondary parties. The value is Y or N.
EFFCTV_RISK_CUTOFF_LVL The effective risk level is specified for the conditional thresholds, which will be decided for overall risk.
ACTVTY_RISK_CUTOFF_LVL The activity risk level is specified for the conditional thresholds, which will be decided for overall risk.
INCLD_ACCT_HLDR_TYP_CD List of Account Types included by the scenario.
MANTAS_BUSINESS_ACCT_TYPES Codes that identify the business purpose or usage of this account for scenarios.
FUNC_CURR_FL

The value is Y or N.

If Y, all aggregation will be done on the functional currency.

Note: If both reporting and functional currency are passed as ā€œNā€, then it will be considered as the base currency.

INCL_WIRE_TRXN_PRDCT_TYPE_LST A list of transaction product type codes for wire transactions is included in the scenario.
INCL_MI_TRXN_PRDCT_TYPE_LST A list of transaction product type codes for monetary instrument transactions is included in the scenario.
INCL_CASH_TRXN_PRDCT_TYPE_LST A list of transaction product type codes for cash transactions is included in the scenario.
INCL_BO_TRXN_PRDCT_TYPE_LST A list of transaction product type codes for back-office transactions is included in the scenario.
LRF_DIGITS Considering the number of the last digit as zero for the round amount.
MIN_TRANS_ROUND_AMT Considering the minimum amount for round amount.
MAX_TRANS_ROUND_AMT Considering the maximum amount for round amount.
MIN_INDIVIDUAL_TRANS_AMT Minimum supported amount for LRT scenario.
STRUCTURED_CASH_LIMIT_MIN Lower limit used to be considered by Financial Institutions
STRUCTURED_CASH_LIMIT_MAX Reporting limit used to be considered by Financial Institutions
DEGREE_OF_PARALLELISM This should be configured properly for performance gain for SQL execution in parallel degree.
For example: model_group_name=VALIDATION, model_name=RMF_LRT, from_date=01-Jan- 2012, to_date=31-Dec-2017, prod_flag=N, include_full_lookback=N, last_run_date=09-May-2016, frequency=7, look_back=30, focus=CUSTOMER, filters=PRIMARY_CUST_FL:Y~INCLUDE_B2B_TRNFR_FL:Y~INCLUDE_TRUSTED_TRANS_FL:Y~I NCL_RLTD_PARTIES:Y~RPTNG_CURR_FL:N~MIN_HRG_RISK_LVL:10~INCL_SEC_PARTY_FL:Y~E FFCTV_RISK_CUTOFF_LVL:10~ACTVTY_RISK_CUTOFF_LVL:10~INCLD_ACCT_HLDR_TYP_CD:C R~MANTAS_BUSINESS_ACCT_TYPES:RBK|RBR~FUNC_CURR_FL:Y~INCL_WIRE_TRXN_PRDCT_ TYPE_LST:EFT-ACH|EFT-TREASURY|EFT-FEDWIRE|EFT-SWIFT|EFTOTHER| EST~INCL_MI_TRXN_PRDCT_TYPE_LST:CASH-EQ-CASHIER-CHECK|CASH-EQ-CERTCHECK| CASH-EQ-MONEY-ORDER|CASH-EQ-TRAVELERS-CHECK|CASH-EQ-OTHER|CASHLETTER| CHECK|PAPER-OTHER|CHECK-ACH~INCL_CASH_TRXN_PRDCT_TYPE_LST:DEBITCARD| SVC|CREDITCARD| CURRENCY|PHYS~INCL_BO_TRXN_PRDCT_TYPE_LST:JOURNAL~LRF_DIGITS:4~MIN_T RANS_ROUND_AMT:10~MAX_TRANS_ROUND_AMT:100000000~MIN_INDIVIDUAL_TRANS_A MT:10~STRUCTURED_CASH_LIMIT_MIN:10~STRUCTURED_CASH_LIMIT_MAX:1000~DEGREE_ OF_PARALLELISM:8
  • Edit Task Parameters and Save.