You are here: Processor Library > About matching processors

About Match Processors

Match processors allow you to match records either from the same source, or from several sources, and to review the results of the matching process.

There are six types of match processor, according to the type of business problem you need to address with matching:

Matching processor

Business problem

Advanced Match

General purpose match processor for any type of matching problem. Allows you to configure matching using all available options.

Deduplicate

Identifying and/or merging duplicate records in a single data set.

Group and Merge

Grouping records in a single data set by an attribute or attributes, and merging groups.

Link

Identifying and linking matching records between data sets.

Enhance

Enhancing an operational data set by matching to one or more reference data sets.

Consolidate

Identifying and/or consolidating duplicate records in a number of data sets, in order to create the "best" records.

These tailored processors guide you through the configuration of matching using default configurations suited to the business problem being addressed.

Within each match processor, the stages of matching are aligned with the real-world logical steps in defining a matching process:

Stage of match process definition	Match Sub-processor
Which data is involved in the matching process?	Input
What are you going to use to identify records?	Identify
How are you going to divide the data into clusters for fast matching?	Cluster
How are you going to compare records, and interpret the comparison results?	Match
How do you want to merge matching records?	Merge

Note that all match processors except Deduplicate and Group and Merge accept more than one source of data. To connect more than one source of data to a process for matching, add a Reader (from the Read/Write family) for each source of data that you wish to involve in matching.

To read more about matching, see the Matching Concept Guide.

Reviewing match results

Reviewing results is a key part of the match process, as one of the key principles of OEDQ is that not every match decision can be automated. Reviewing match results manually allows users to apply specific intelligence to individual cases, thereby validating and refining the match process.

OEDQ provides two review applications - Match Review and Case Management. These applications divide up the match results in different ways, so match processors must select, in advance, which review application will be used to review their results. Once a match processor has been defined and executed, its results can be reviewed by the configured application.

For more details on the available review options, see Reviewing matching results.