1.3.11.42 Transliterate

The Transliterate processor converts strings from one writing system (such as Arabic) to another (such as Latin). This is a largely phonetic operation which attempts to create an equivalent of the original string in the target writing system, based on the sounds that the string represents. No attempt is made to translate the string. For example, the Arabic string which sounds like 'bin' when read aloud and which is a common component of Arabic names is transliterated to the Latin string "bin", not translated to its literal meaning, 'son of'.

Note that a single string in the original writing system may have several valid transliterations. For example, 'bin' may also be transliterated as 'ben'. Some names may have very many alternate transliterations. The Transliterate processor aims to provide a single, standard form of the original string, not all the possible alternative transliterations. Instead, alternative transliterations are recognized as part of the matching process, where it is managed in a similar way to recognizing alternative spellings of non-transliterated names.

The EDQ Transliterate processor is built around the ICU4J libraries provided by ICU. ICU is released under a nonrestrictive open source license that is suitable for use with both commercial software and with other open source or free software. For more information about ICU and the ICU license, visit the ICU website.

Use the Transliterate processor to convert strings in a phonetically appropriate manner from one writing system to another. This is useful when matching strings provided in one writing system against reference data that is provided in a different writing system. For example, international watch lists are often provided only in Latin script.

Note:

The Transliterate processor is not the only available tool for handling alternate writing systems in EDQ. Depending on the complexity of the transliteration requirements and the support for the various writing systems in ICU4J, other approaches may be more reliable. For example, it is possible to implement transliteration using a combination of the Replace and Character Replace processors, along with a suitable set of reference data for the source and target writing systems.

The following table describes the configuration options:

Configuration	Description
Inputs	Specify any number of String attributes, or arrays of String attributes, that you want to transliterate. There is no need to transliterate Number or Date attributes, as they are stored in a format which is independent of any particular writing system. Strings containing numbers or dates will be converted to the target writing system in the most appropriate fashion, but this is not a phonetic operation.
Options	Specify the following options: List of possible transliteration options: defines the source and target writing systems to be used in transliterating the input. Specified as a standard list resource Default value: Any to Latin.
Outputs	Describes any data attribute or flag attribute outputs.
Data Attributes	The following data attributes are output: Transliterated: the version of the attribute value, transliterated into the target writing system.
Flags	None.

The Transliterate processor does not output any summary data. The transliterated input value is displayed with the input attributes in the data view.

Output Filters

None.

Example

For example, names in the input data can be transliterated from Greek ("Original Script Name") to Latin ("Original Script Name.Transliterated"):