Skip to content

Value normalization overview

The values of extracted fields and tag instances can be the original text that triggered the condition of the extraction and/or tagging rule or something else that is derived from the text with a process called normalization. The aim of such a process is to remove or reduce redundancies in the data, to present it in a uniform or standardized format and to improve completeness and correctness.
For example, original text:

I earned forty dollars.

Extracted (normalized) value:

40 dollar

Normalization can be automatic or optional. Default normalization is performed automatically, while transformation and composition are optional. Optional normalization techniques can be used separately or in combination.