One endpoint for multilingual detection and protection with automatic language/country context, stable customer-facing entity types and structured detection metadata.
37 · 153 · 52
languages · country profiles · strong-format country profiles
Language packs include English, Urdu, Arabic, Hindi, Indonesian, Filipino, Persian, Turkish, Bengali, Swahili, Chinese, Japanese, Korean, Russian and 23 more.
The supplied taxonomy contains 153 country profiles: 52 strong-format profiles plus 101 baseline-contextual profiles.
The supplied taxonomy contains 1,599 configured common, fallback, language and country recognition rules.
PrivoNest can infer multiple active languages and country candidates, then expand relevant language packs from selected country metadata.
Checksums and validators cover identifiers such as IBAN, Luhn cards, Aadhaar/Verhoeff, CPF/CNPJ, national IDs and additional local formats.
Responses can include type, recognizer, span, confidence, source, language, country and validator without returning the raw matched value.