Match Transformation: Make Array from String

1.3.4.9.14 Match Transformation: Make Array from String

The Make Array from String transformation allows a single text value to be broken up into a variable number of distinct values. This is useful when creating clusters for matching, as clusters will be created for each distinct value created. This ensures that any values with a common word in them, regardless of the order of that word within the value, will be in the same cluster for matching purposes. For example, where a Name identifier has values 'John Simpson' and 'Simpson, J', clustering by making an array using comma and space delimiters will ensure the two records are in the same cluster ('Simpson').

The Make Array from String transformation is functionally the same as the main Make Array from String processor, but is used specifically when clustering to split values into several words to use as cluster keys.

Note that Make Array from String cannot be used within comparisons.

Use the Make Array from String transformation as the last transformation when clustering in order to ensure that records will be brought together into the same cluster if they have any word in common.

The following table describes the configuration options:

Configuration Description

Configuration	Description
Options	Specify the following options: `Delimiter Reference Data`: provides a way of specifying a standard, reusable set of delimiter characters or Strings for breaking up data, and allows you to use control characters as delimiters. Type: Reference Data. Default value: *Delimiters. `Delimiter characters`: allows you to specify delimiters to use without having to create reference data for simple delimiters such as space or comma. Note that if these are used in addition to a reference list, all delimiters from both options will be used to break up the data. Type: Free text. Default value: Space.

Options

Specify the following options:

Delimiter Reference Data: provides a way of specifying a standard, reusable set of delimiter characters or Strings for breaking up data, and allows you to use control characters as delimiters. Type: Reference Data. Default value: *Delimiters.
Delimiter characters: allows you to specify delimiters to use without having to create reference data for simple delimiters such as space or comma. Note that if these are used in addition to a reference list, all delimiters from both options will be used to break up the data. Type: Free text. Default value: Space.

Example

In this example, the Make Array from String transformation is included in the configuration of a cluster on an Address1 identifier.

Example configuration

The following transformations are added to the Address1 identifier to form a cluster:

Upper Case
Strip Numbers
Strip Words (to remove very common words such as The, House, Road, Street, Avenue, Lane, etc.)
Normalize Whitespace
Make Array from String

Example transformations

The following table shows examples of transformations using the above configuration:

Table 1-86 Example Transformations for Make Array from String

Value	Value after first 4 transformations	Value after Make Array from String transformation
The Maltings, 14 Appletree Lane	MALTINGS, APPLETREE	1 - MALTINGS 2 - APPLETREE
14 Appletree Lane	APPLETREE	1 - APPLETREE
The Maltings	MALTINGS	1 - MALTINGS
32 Rushton Road, Coventry	RUSHTON, COVENTRY	1 - RUSHTON 2 - COVENTRY
32 Rushton Rd	RUSHTON	1 - RUSHTON
15 Stroud Green Road	STROUD GREEN	1 - STROUD 2 - GREEN
14 Green End Avenue	GREEN END	1 - GREEN 2 - END

All records that share a common value after transformation will be in the same cluster. For example, the first two records above will be in the 'APPLETREE' cluster, and the first and third records will be in the 'MALTINGS' cluster.