Person name matching in Grid
The Moody's screening logic for entity names classified as Persons evaluates each inquiry submission to determine its likely cultural and linguistic origin.
Person entity name matching logic
For person name matching, Moody's uses an analytical-linguistic identity matching approach to overcome the challenges faced by international firms. The Grid name matching engine leverages linguistic and onomastic intelligence to do the same for automated name matching.
For all inquiry searches on names identified as a Person, the Grid name matching engine:
Analyzes and diagnoses the overall “linguistic culture” of a name's origins. Factors such as the Romanization and changed order of names are examined at this time.
Rearranges and compares each part of a name. Using culturally sensitive logic, each name is parsed to differentiate first names from surnames, family names, titles, and so on.
Recognizes and accounts for cultural and language differences related to character order, diphthong combinations, and consonant strings.
Identifies a name's gender and generates alternative spellings based on phonetics and transliteration differences.
Determines the most likely country of association of the name.
Note
For best results, each person’s name should be submitted in the order of its culturally relevant naming convention.
Match results
Exact name match: Produced when a submitted name matches an entity name or alias exactly, letter-for-letter.
This comparison is not case-sensitive and ignores diacritical marks and special characters. An exact match generally produces a match score of 99 or 100 before rescoring.
Variant based name match: Based on linguistic variants and non-linguistic variants.
Linguistic variants: Based on known linguistic variants of a name part, which applies to alternative or derivative spellings of a name, phonetic and transcription variants, nicknames, and hypocorisms. Linguistic variants include transcription variants, homophones, diminutive, and cultural variants of the same name.
Non-linguistic variants: Include matches based on misspellings, typographic errors, or name omissions. The match score resulting from non-linguistic name variations can vary widely and depend on the type and severity of the differences and the cultures of the names involved.
Non-linguistic variants include matches based on initials, missing or extra name parts, name order differences, misspellings, different fields in the data structure, OCR (optical character recognition) scanning errors, and typographic errors such as letter omission, insertion, or transposition.
Compressed name match: Produced when names match only when all white spaces are removed from names.