The data cleaning feature in data product configuration enables you to replace known bad values in your source data with null, when Tamr finds an exact, case insensitive match for an attribute value. This helps to ensure that these values are not used for matching or included in your golden records

In addition to values you specify for data cleaning, Tamr automatically replaces the following values with null wherever it identifies an exact, case insensitive match in the mapped source record fields:

ABSENT
BLANK
CC ONLY DO NOT INACTIVATE
CCCCCCCCCCC
CCCCCCCCCCCC
CCCCCCCCCCCCC
COPYFROM
COPY FROM
DO NOT INACTIVATE
DO NOT DEACTIVATE
EMPTY
INVALID
MISSING
N/A
NA
NAN
NIL
NO CO SPECIFIED
NO DATA
NO INFO
NO STREET
NONE
NONEXISTENT
NOT APPLICABLE
NOT AVAILABLE
NOT FOUND
NOT PROVIDED
NOT PROVIDED ACCOUNT - UNKNOWN
NOT SET
NOT VALID
NULL
[email protected]
STREET
TBD
UNAVAILABLE
UNDEFINED
UNKNOWN
VACANT
[[UNKNOWN]]
[UNKNOWN STREET]
[UNKNOWN STREET)
UNKNOWN VALUE
UNKNOWN ZIP
UNRESOLVED
UNSET
UNSPECIFIED
VOID

Healthcare Provider NPI Cleaning

In the Healthcare Provider data product, Tamr replaces the following invalid National Provider Index (NPI) values with null:

NPIs that include 5 or more of the same digit repeated
NPIs that contain only zeros
NPIs that contain the substring 123456789

Additionally, Tamr removes any decimal zeros. For example, if the value is 555.000, Tamr changes the value to 555.

Example

Consider this example in which Tamr automatically replaces attribute values of STREET will null

The following attribute values are replaced with null:

street
Street
STREET

The following values are not cleaned; the full value remains in the attribute:

streets
123 main street
ELM STREET
St.