Skip to main content

Rules

Rules are the fundamental building blocks of the PredictaID Decision Engine. A Rule defines a specific condition and the consequence (Effect) if that condition is met.

Structure​

A Rule consists of:

  1. Conditions: Logic to evaluate (e.g., Age < 18, Document exists in Blocklist).
  2. Effect: What happens when constraints are met.
  3. Severity: Importance of the rule (LOW, MEDIUM, HIGH, CRITICAL).

Conditions​

Conditions compare a Field against a Value using an Operator.

  • Field: Data from the transaction (e.g., transaction.amount) or enriched data (e.g., bigData.age).
  • Operators: EQUALS, GREATER_THAN, MATCHES_REGEX, DATE_DIFF, etc.

Effects​

  • SCORE: Adds or subtracts points from the final risk score.
  • BLOCK: Immediately triggers a rejection or requires manual review.
  • FLAG: Adds a warning flag to the profile without altering the score directly.

Dynamic Data​

Rules can access external data providers (e.g., BigDataCorp, Receita Federal) during evaluation, allowing for complex checks like:

"If CPF Status is SUSPENDED AND Transaction Amount > 5000, THEN BLOCK."