Cert status normalize
Exacthealthcare_cert_status_normalizeDeterministic and authoritative — the canonical answer for this input.
Normalize a raw certification status string to the canonical lifecycle: active, pending, not_certified, expired, revoked, or unknown. Exact synonyms ('Board Certified', 'Inactive', 'Lapsed') match at confidence 1.0; messy variants ('Cert Pending', 'revoked for cause', 'lapsed 2019') match via a deterministic keyword pass at confidence 0.8. Unrecognized input returns 'unknown' at confidence 0 — never a guess. Processing multiple inputs? Use `healthcare_cert_status_normalize_batch` to do up to 1000 in one call.
Parameters
valuestringrequiredRaw certification status string.
Call it
curl -X POST https://api.snipget.ai/v1/healthcare/cert-status/normalize \
-H "Authorization: Bearer sk_live_..." \
-H "Content-Type: application/json" \
-d '{"value":"Board Certified"}'The Batch tab sends up to 1000 inputs in one call via /v1/healthcare/cert-status/normalize/batch.
Example response
{
"input": "Board Certified",
"status": "active",
"matched": true
}FAQ
What problem does this solve?
Source data describes certification status a hundred different ways: 'Board Certified', 'C', 'lapsed', 'eligible'. This maps each one to a fixed lifecycle so you can filter and compare them reliably.
What are the possible outputs?
One of six canonical values: active, pending, not_certified, expired, revoked, or unknown. Matching is deterministic: an exact synonym map first (confidence 1.0), then a keyword scan for messy variants like 'Cert Pending' or 'revoked for cause' (confidence 0.8). The same input always returns the same status.
What if it does not recognize my value?
It returns status 'unknown' with matched false and confidence 0, rather than guessing. Check the confidence: 1.0 means an exact synonym, 0.8 means a keyword match worth a glance, 0 means unmatched.
Does it use dates?
No. This works purely from a status word or phrase. If all you have is an effective and expiration date, use the infer-from-dates tool instead.