NULL_OR_BLANK Match
Changes how Zingg handles null and blank values in a field.
What NULL_OR_BLANK does
By default, most Zingg match types treat a null or blank value as automatically matching anything.
NULL_OR_BLANK adds a separate feature flagging whether both values were actually present, so the model can learn to discount matches that were really just two missing values.
Exception: NUMERIC does not follow the "null matches anything" default - a null paired with any value scores as a non-match under NUMERIC, the opposite of every other match type covered here.
Use NULL_OR_BLANK combined with another match type, in the matchType string with a comma-separated combination.
What NULL_OR_BLANK matches and what it does not
Value A
Value B
NULL_OR_BLANK feature
Notes
null
John Smith
0.0
One side null
[empty string]
John Smith
0.0
Empty string treated exactly like null
null
null
0.0
No distinction between one-sided and both-sided nulls
null
[empty string]
0.0
Null and blank are interchangeable
John Smith
John Smith
1.0
Both values present
There is no distinction between "both null" and "one null, one populated" - all four null/blank rows above score identically.
When to use NULL_OR_BLANK
When not to use NULL_OR_BLANK
Last updated
Was this helpful?