Senzing Globalization Guide

What languages does Senzing support?

Senzing utilizes UTF-8 encoding which allows for most languages of the world to be properly captured and processed. Beyond ingesting and storing data, Senzing analytics go further – taking into consideration domain, culture, and cross-script differences for comprehensive global entity resolution. Senzing provides native support for cross-script comparisons across many languages and writing systems, with entity-centric learning capabilities that allow it to discover attribute variations (including script variations) even when attributes cannot be matched in their original forms.

Advanced personal name comparisons

Supported cultural groups

Personal names present unique challenges in global entity resolution. Senzing leverages IBM's InfoSphere Global Name Management for culturally-aware name comparison. This world-class name library uses spelling patterns and country-of-association information to determine cultural provenance and optimize matching strategies.

Primary Cultural Groups

Southwest Asian

European

Han

Additional Cultural Support

Culture identification for shared scripts

New in v4.4Starting in v4.4, Senzing distinguishes name cultures that share a script, so each name is handled with the rules for its actual culture. For example, Ukrainian names in Cyrillic are Romanized with Ukrainian rules rather than generic East Slavic ones.

Organizational names

Senzing provides robust same-script organizational name matching across many writing systems and languages.

Same-script organizational name matching examples

CJK+English cross-script matching

New in v4Senzing v4 introduces native cross-script matching between CJK (Chinese, Japanese, Korean) and English organizational names without requiring reference data.

For other cross-script language combinations, robust matching of organizational names may still require reference data containing multiple versions of names, as there is no consistency in how organizations handle name translation across scripts. Some organizations represent names phonetically (Romanize them), some translate (or translate parts), and some organizations rebrand when moving into new markets/scripts. For these scenarios, data providers or services that offer organizational name enrichment can be beneficial.

Romanized Burmese organization names

New in v4.4Starting in v4.4, common Romanized Burmese organization terms match their English equivalents:

Enhanced address comparisons

New in v4Senzing provides cross-script matching capabilities for addresses. Starting in v4, native cross-script matching between CJK (Chinese, Japanese, Korean) and English addresses is supported without requiring reference data, representing a major improvement for global address resolution.

Address matching examples

New in v4.4Matching Japanese block-style addresses (丁目番号 ↔ 1-1-1), as in the Tokyo example, requires v4.4 or later.

For other language combinations, addresses can be challenging for entity resolution as they tend to have many data quality issues. Senzing has capabilities to handle native scripts for addresses, with the most effective processing occurring in native-to-native (rather than native-to-Romanized) scenarios.

For cross-script address comparison in non-CJK languages, using an address hygiene product to Romanize addresses and providing both native script and Romanized versions to Senzing can improve matching accuracy.

Getting started

Senzing provides comprehensive globalization capabilities out of the box. The breakthrough CJK+English cross-script matching capabilities introduced in v4 for both organizational names and addresses require no additional configuration - simply upgrade to v4 to benefit from these improvements.

For the rare cases requiring specific cultural tuning or advanced cross-script scenarios, contact Senzing Support for guidance on optimizing configurations for your specific use cases.