Fuzzy Grouping Transformation Editor (Advanced Tab)
Applies To: SQL Server 2016
Use the Advanced tab of the Fuzzy Grouping Transformation Editor dialog box to specify input and output columns, set similarity thresholds, and define delimiters.
To learn more about the Fuzzy Grouping transformation, see Fuzzy Grouping Transformation.
Input key column name
Specify the name of an output column that contains the unique identifier for each input row. The _key_in column has a value that uniquely identifies each row.
Output key column name
Specify the name of an output column that contains the unique identifier for the canonical row of a group of duplicate rows. The _key_out column corresponds to the _key_in value of the canonical data row.
Similarity score column name
Specify a name for the column that contains the similarity score. The similarity score is a value between 0 and 1 that indicates the similarity of the input row to the canonical row. The closer the score is to 1, the more closely the row matches the canonical row.
Set the similarity threshold by using the slider. The closer the threshold is to 1, the more the rows must resemble each other to qualify as duplicates. Increasing the threshold can improve the speed of matching because fewer candidate records have to be considered.
The transformation provides a default set of delimiters for tokenizing data, but you can add or remove delimiters as needed by editing the list.