Decision Scoring API Use Case
Problem
Section titled “Problem”A match score alone is not enough to decide whether a workflow should automate, review, or reject a result.
How AvelinLabs Helps
Section titled “How AvelinLabs Helps”AvelinLabs returns decision-support fields that help applications interpret result strength. Confidence, trust, uncertainty, ambiguity, and quality fields can be used to route outputs according to a product’s own policy.
Why It Is Different
Section titled “Why It Is Different”Avelin separates the proposed result from the strength and limits of its evidence. Review and ambiguity labels make human attention an explicit workflow outcome rather than hiding uncertainty inside one score.
Input Signals
Section titled “Input Signals”Decision scoring may consider signals such as:
- result confidence
- trust or quality indicators
- uncertainty and ambiguity
- strength of the top ranked result
- supporting skill or occupation evidence
Decision Output
Section titled “Decision Output”The current beta documentation describes decision labels such as:
AUTO_ACCEPTREVIEWREJECTAMBIGUOUS
These labels are intended to help downstream systems decide what should happen next. They should be combined with product-specific review and risk policies.
AUTO_ACCEPT
Section titled “AUTO_ACCEPT”- Meaning: Strong enough for the configured workflow.
- Typical action: Use or display automatically where risk is low.
REVIEW
Section titled “REVIEW”- Meaning: Useful but needs human review.
- Typical action: Send to a reviewer or consultant.
REJECT
Section titled “REJECT”- Meaning: Not reliable enough.
- Typical action: Do not use for automation.
AMBIGUOUS
Section titled “AMBIGUOUS”- Meaning: Multiple interpretations or weak evidence.
- Typical action: Ask for more input or review manually.
Product Fit
Section titled “Product Fit”Decision scoring can support analyst review queues, workflow routing, quality control, dashboards, and human-in-the-loop automation.
GitHub Example Reference
Section titled “GitHub Example Reference”Complete executable examples should live in the official AvelinLabs API examples repository.