Developer Transformation Guide

10.5
- 10.5.2
- 10.4.1
- 10.4.0

Back Next

Reference Data Use in the Labeler Transformation

Informatica Developer installs with different types of reference data objects that you can use with the Labeler transformation. You can also create reference data objects.

When you add a reference data object to Labeler transformation strategy, the transformation searches the input data on the strategy for values in the reference data object. The transformation replaces any value it finds with a valid value from the reference data object, or with a value that you specify.

The following table describes the types of reference data you can use:

Reference Data Type	Description
Character sets	Identifies different types of characters, such as letters, numbers, and punctuation symbols. Use in character labeling operations.
Probabilistic models	Adds fuzzy match capabilities to token label operations. The transformation can use a probabilistic model to infer the type of information in a string. To enable the fuzzy match capabilities, you compile the probabilistic model in the Developer tool. Use in token labeling operations.
Reference tables	Finds strings that match the entries in a database table. Use in token labeling and character labeling operations.
Regular expressions	Identifies strings that match conditions that you define. You can use a regular expression to find a string within a larger string. Use in token labeling operations.
Token sets	Identifies strings based on the types of information they contain. Use in token labeling operations. Informatica installs with token sets different types of token definitions, such as word, telephone number, post code, and product code definitions.