High-quality address data is a critical part of doing business.
Organizations that do business in multiple countries or regions must
maintain the quality of address data across different postal standards
and languages. IBM® InfoSphere® QualityStage® Address Verification
Interface includes
a stage that provides processes that can organize, verify, and transform
address data.
After you install and configure the product, an Address Verification
stage is available to use in
IBM InfoSphere DataStage® and QualityStage Designer.
Use the Address Verification stage to process address data in one
of the following ways:
- Parse
- Identifies address elements. You parse address data in preparation
to standardize the format of the address.
- Validation
- Checks each address element for accuracy against postal validation
reference files and then changes, improves, or completes each address
element, if necessary and possible. You validate address data to ensure
that each address conforms to postal standards and to assess the deliverability
of each address.
Transliteration
Transliteration is available
with both the validation and parse processes. Transliteration is the
process of representing text from one script in another script. Transliteration
is not translation. You use transliteration to take records that are
in multiple scripts and represent the records in a common script.
You can then work on the records as one group. An example of transliteration
is converting addresses in Hebrew or Chinese script to Latin script.
Another example is shown in the following table.
Table 1. An example
of transliteration shown with Chinese and Latin scriptsRepresentation |
Example |
Chinese script |
|
Translation in Latin script |
Street |
Transliteration in Latin script |
Jie |
Transliterated address data is useful because it can
produce addresses in a native script or produce a common representation
of different scripts. A common representation enables a wider audience
to understand, work with, and standardize data.