Solution - trillium_quality - trillium_discovery - Latest

Trillium Beyond The Basics Guide

Product type
Software
Portfolio
Verify
Product family
Trillium™ software
Product
Trillium™ software > Trillium™ Quality
Trillium™ software > Trillium™ Discovery
Version
Latest
ft:locale
en-US
Product name
Trillium Quality and Discovery
ft:title
Trillium Beyond The Basics Guide
Copyright
2024
First publish date
2008
ft:lastEdition
2024-10-18
ft:lastPublication
2024-10-18T15:32:19.232000
L1_Product_Gateway
Verify
L2_Product_Segment
Data Quality
L3_Product_Brand
Precisely Trillium
L4_Investment_Segment
Legacy DQ
Core Data Quality
L5_Product_Group
Legacy DQ - Application
Data Quality - Application
L6_Product_Name
Trillium Quality
Trillium Discovery

Trillium Quality enables you to include multi-matching processes in a project. Multi-matching is a process for more complex matching. You define multiple window keys, one for each matching criteria, and run multiple Relationship Linkers on the same set of records using the different window keys. For example, the standard window key is based on name and address data. You could create a second window key that is based on a more constant attribute, such as an e-mail address or social security number.

When multiple Relationship Linkers are run, the linking results from each process are recorded in a link file, and then transitivity of the links are resolved using the Resolve utility.Transitivity occurs when two records are linked together indirectly through a third record. For example, record A may have linked to record B in the first matching and record B may link to record C in a subsequent matching. When this happens, record A has linked to record C through transitivity. The Resolve utility will find all indirect links and reconcile the linking relationships. Resolved results are then applied to the data by the File Update utility and the final matching is performed.

The following pages present a sample project that uses three matching criteria: a matching based on name and address, e-mail address, and phone number. Your solution will depend on the data that your records contain, but basic procedures apply to any multi-matching projects.