How The Advisor™ Rule Engine Works (1 of 3)
There are two techniques for progressing toward a conclusion in a rule-based system:
- Data driven or forward rule evaluation:
- start with known data and progress naturally to the conclusion
- e.g. given a request for quote, what insurance premium should we quote?
- Look forward from request’s data (age, location, requested coverage, prior claims, ...)
Folie 48
Nächste Folie
- Goal driven or backward rule evaluation:
- select a possible conclusion and try to prove its validity by looking for supporting evidence
- e.g. can we approve the automobile rental?
- Yes, but only if driver has valid, no violations license
- Look back and test for a valid license
- Look back and test for no violations license