Working with Matching Nodes
Matching refers to the process of comparing nodes in various contexts to either identify nodes that are the same and merging them to prevent duplication of data or to map source nodes to target nodes in a mapping viewpoint.
Use Cases
The following table illustrates the different use cases for matching in Oracle Fusion Cloud Enterprise Data Management:
Table -32 Matching Use Cases
| Use Case | Definition | Example |
|---|---|---|
| Merge Request Items | Match incoming nodes in a request to existing nodes in a viewpoint in order to identify and merge nodes that are the same. See Understanding Matching and Merging Nodes. | You have a Product viewpoint and you are bringing in data from a new supplier and merging it into an existing standard parts catalog. Use a match and merge operation to match incoming nodes in a request with product nodes that already exist in the viewpoint and then merge the records together so that you do not duplicate part data in your viewpoint. |
| Map Request Items | Match nodes in a source node type to nodes in a target node type and map them together by creating a parent-child relationship between them in a hierarchy mapping viewpoint. See Understanding Matching and Mapping Nodes | You have a legacy chart of account viewpoint that you are mapping to cloud GL accounts. Use a match and map operation to match incoming legacy nodes in a request with their corresponding cloud accounts and map them in a hierarchy mapping viewpoint. Note that nodes are not merged together as a result of a mapping operation. |
| Deduplicate Viewpoint | Match existing nodes in a viewpoint in order to identify and merge nodes that are the same. See Understanding Deduplication. | You have a Customer viewpoint with customer information from different data sources, and as a result some nodes are duplicates of one another. Use a deduplicate viewpoint operation to match different nodes that represent the same customer and merge them together into a single record. |
Terminology
The following terms can help you understand the matching process:
- Data source: An object that represents the source for the incoming data
to be matched and linked in Cloud EDM. This can
be either another Cloud EDM application
(called a registered data source) or an external system whose data is not
being managed in Cloud EDM (called an
unregistered data source). See Understanding Data Sources.
Note:
You can merge and map request items for any data source. You can deduplicate data for registered data sources only. - Matching rule: Controls how nodes are matched either from an incoming data source to nodes that already exist in a node type (for merging and mapping) or in a viewpoint (for deduplication). See Creating, Editing, and Deleting Matching Rules.
- Survivorship rule (Merging and Deduplication only): Specifies which properties and relationships from the source node get merged into the target nodes in a viewpoint after a match has been confirmed. See Creating, Editing, and Deleting Survivorship Rules.
- Matching workbench: Enables you to review match candidates based on the criteria from the matching rules and accept the ones that you want to merge into the existing nodes. See Matching for Merging, Mapping, or Deduplicating Nodes.
- Clustering property (Deduplication only): A property that you identify to group nodes into clusters so that you can run matching on them in order to identify and combine duplicate nodes.