Examining Duplicate Data Rows and Values - trillium_discovery - 17.1

When you import data into a repository, the data is analyzed to identify structural information such as source type, minimum and maximum row length, the count of duplicate values, and the count of rows that are potentially a duplicate of other rows in an attribute. Duplicate data rows indicates non-unique values. Review duplicate values and duplicate data rows to identify candidates for remediation.

To examine potential duplicate data rows

1. Open a data source.
2. Click the Source Metadata tab.
3. Double-click Duplicate Rows if the Value column shows a number greater than 0. The Duplicates tab opens showing the following metadata for each potential duplicate value:

   Column Name	Description
   Duplicate ID	Indicates whether a data row is a duplicate of one or more data rows in the data source. A value of 1 indicates a data row is in the first group of duplicates found, a value of 2 indicates a data row is in the second group of duplicates found, and so on. Data rows with a value of 0 are unique.
   Duplicate Value Frequency	The number of times the duplicate value occurs.
   Distribution %	The measure of how much the attribute contains the duplicate data value.

4. Examine the Duplicate Value Frequency column. The value indicates the number of data rows that contain values that are potential duplicates. Attributes can have one or multiple potential duplicate values. Each is assigned a unique Duplication ID value.
5. Double-click a row whose values you want to examine. The Data Rows: Duplicate ID value tab opens showing the rows that contain the duplicate value. Examine the rows to verify that the information is or is not a duplicate.
   Note: If the rows contain no values, then the shared duplicate value is a null (empty) value.
6. Optional: Export selected rows to your local or server system as a .CSV file. For more information, see Exporting Tab Rows.