OFAC Name Screen Threshold Scorer

Bounded Levenshtein name similarity against independently resolved OFAC reference with caller review cutoff, normalized names explicit; not sanctions clearance, identity match or verified sanctions list. Caller independently verifies edition, jurisdiction, applicability and one currency; screening does not establish legal compliance, coverage or bank approval.

Description

Name similarity measures differences between strings under a chosen normalization convention.

Name similarity measures differences between strings under a chosen normalization convention.^[1]

Similarity alone does not establish identity or a sanctions-list match. Use OFAC Name Screen Threshold Scorer to evaluate the supplied inputs under the model explained below.

When to use OFAC Name Screen Threshold Scorer

Use OFAC Name Screen Threshold Scorer to evaluate the supplied inputs under the model explained below. Establish the currency and minor-unit convention, measurement period and applicable transaction or institution context.

How to use OFAC Name Screen Threshold Scorer and understand its results

Establish the currency and minor-unit convention, measurement period and applicable transaction or institution context.
Enter Caller normalized candidate names, Caller resolved ofac name, Caller review similarity threshold using the field table and validation rules below, then run the tool.
Read Observed name similarity scores, Caller name review candidates; interpret the units, signs, ordering and unknown-value placeholders described below.
Correct reported input errors and rerun. Apply successful results according to their specific meanings and check the assumptions before using them in a decision.

Input field	Meaning	Value type	Requirement
callerNormalizedCandidateNames	Caller normalized candidate names	array of string	Required; constraints below
callerResolvedOfacName	Caller resolved ofac name	string	Required; constraints below
callerReviewSimilarityThreshold	Caller review similarity threshold	number	Required; constraints below

Enter an independently resolved nonblank reference name of at most 128 UTF-16 code units, nonblank normalized candidate names and a similarity threshold from 0 to 1. Compare strings exactly using Levenshtein insertion/deletion/substitution distance on UTF-16 code units, then score 1−distance/max(lengths). Return every score and candidates at or above threshold in original order. An empty candidate array returns empty results. No normalization, OFAC list retrieval, identity resolution or official screening algorithm is performed.

similarity=1−LevenshteinDistance/max(candidateLength,referenceLength); review=score≥threshold

Use the field order, units and branch conditions described above.

Output field	Meaning	Value type
observedNameSimilarityScores	Observed name similarity scores	array of number
callerNameReviewCandidates	Caller name review candidates	array of string

json

{
  "input": {
    "callerNormalizedCandidateNames": [
      "ALICE",
      "ALICE B"
    ],
    "callerResolvedOfacName": "ALICE",
    "callerReviewSimilarityThreshold": 0.8
  },
  "output": {
    "observedNameSimilarityScores": [
      1,
      0.7142857142857143
    ],
    "callerNameReviewCandidates": [
      "ALICE"
    ]
  }
}

Limitations and assumptions

Similarity alone does not establish identity or a sanctions-list match. Apply the exact scope and assumptions explained above.

All listed fields require explicit values. Unexpected fields, incorrect types, nonfinite numbers and sparse arrays fail validation. Domain-specific bounds, alignments and supported tokens are explained above; extremely large inputs or outputs can also be rejected. Input errors differ from successful false results, empty match lists and unknown-evidence placeholders.

Alternative or Complementary methods

Use Bust Out Sleeper Fraud Limit Maximize Vanish Pattern Lookup for a related task: Bust-out descriptions concern reported fraud patterns over time.

Use App Fraud Authorised Push Payment Uk Crm Reimbursement Code Lookup for a related task: APP fraud reimbursement depends on jurisdiction, payment scheme, facts and effective rules.

References

OFAC Name Screen Threshold Scorer — tool.io

Similar or alternative tools

Bust Out Sleeper Fraud Limit Maximize Vanish Pattern Lookup
Bust-out descriptions concern reported fraud patterns over time.
App Fraud Authorised Push Payment Uk Crm Reimbursement Code Lookup
APP fraud reimbursement depends on jurisdiction, payment scheme, facts and effective rules.