4. Simple Matching
The simple matching interface is intended to allow locally held lists of people to be matched against CUD data. This could provide the following benefits:
- Discover whether a record for a person exists in a central data source
- Discover the affiliations recorded for a person
- Enrich locally held data with additional attributes
-
Aid with lifecycle management by exposing when affiliations are no longer valid
The Simple Matching display a list of matching operations previously executed (none in the screen shot below), and a button to create a new matching list.
Here we create one called “Training Demo”
and click to manage it from the list
A blank matching list is now displayed.
To start we upload a file for matching
The column headings must be left as they are in the sample. Add as much data as you are able. Please note that the fields that can be provided are limited, but this is likely to be extended during the early adopter phase.
We are asked to wait while the file is processed
Returning to the matching list we can see which rows matched data in CUD, the number of matches made for each of these, and which rows did not match any data.
Where a match has been made it (or they) can be reviewed by clicking the icon on the row in question.
At which point it is possible to see on what basis that match was made, and the level of confidence that has been assigned to the matching strategy.
Multiple matches can be reviewed and false matches removed
The resulting list can be downloaded as a csv file containing the data you uploaded and data matched from CUD
The matching list will remain in CUD for up to one year, but the matching data is a snapshot of what CUD contains when the match is made and so is not updated as data changes in CUD.
Up: Contents Previous: 3. Simple Searching Next: 5. Affiliations Manager