When adding a data source to the Discovery Center you have the option to add new, custom attributes to the data source. You can also add a new attribute column to some tab views where the Choose Column option is available. You build the attributes and columns using the Expression Builder functions, operators, and attributes in the associated data source.
- Guidelines for adding attributes to data sources:
-
- The attribute will be added to the Data Preview window as a new column. This becomes a permanent attribute that remains in the data source and treated as any other attribute for subsequent review, including use in business rules.
- Before you add the attribute, you should be familiar with the data you are loading into the data source.
- Give the attribute a name that represents the results of the data value or the function that you will be comparing or testing the expression against. The attribute name must be unique; you cannot add two attributes with the same name in a single data source.
- You can edit the attribute name and description at any time after you save the attribute.
- Learn more about adding a new attribute:
-
New attributes added to a data source become fully profiled attributes. The following are benefits and use cases to consider when adding attributes to data sources:
- Combining columns. For example, combine the first, middle, last name attributes into a single attribute to better see complete name data.
- Calculating new information based on existing attributes in the data source. For example, if you are working with business data and you want to see the product stock you currently have available, you could create an attribute named Available and write an expression that subtracts the products ordered from the products on hand: ([On Hand] - Ordered). The resulting values display in a column named Available for all rows of products in the data source.
- Using concatenated keys to create/discover joins. If you have one data source that uses multiple attributes as a composite key and another that has a single key comprised of multiple elements, you cannot create a join between the two. Adding an attribute with this combined expression will allow you to discover/create joins.
- Creating flags. Based on a filtered condition, you can set literal flags relating to given rows that contain the condition. For example, If the income attribute for a customer record is >100k, you could set a literal = "high value". You could then use that as a filter when reviewing profiling results or as part of a business rule expression.
- Assessing compound data for duplicates. To evaluate duplicate data elements (for example, name and address or selected address elements), you can build a new column for those attributes to see an overview of duplicate data.
- Evaluating data segments to provide visual insight into data content and structure. For example, create an attribute that includes the first three digits of postal codes or specific month, day, or year information. This allows you to see the profiled contents on the added column such as nulls, duplicates, and value distribution. You can create recode tables on these values and use them in business rules, which helps simplify the process of writing other business rules.
- Guidelines for adding attribute columns to tab views:
-
- When you add a column to the Data Rows tab of a data source, this new user-defined attribute column is active and displays each time you open the Data Rows tab for that data source, but is not analyzed or added as a permanent attribute to the data source.
- You can add an attribute column for the purpose of viewing business rule
results. Tip: Add an attribute column that will contain the results of a rule expression. Give the column a name that represents the results of the data value or the function that you will be comparing or testing the rule against. For example, if you are working with business data and you want to see the product stock you currently have available, you could create a column named Available and write a business rule expression that subtracts the products ordered from the products on hand: ([On Hand] - Ordered). The resulting values display in an attribute column named Available for all rows of products in the data source.
- When you add an attribute column to any tab view other than Data Rows, the new column is visible only for the period in which the tab view remains displayed. When you close the tab, you lose the column and its data.
- You can edit the new attribute column at any time.
- The following tab views do not support adding a custom attribute column:
- Metadata: Selected Join
- Tasks in Queue from the Task Manager
- Source Metadata. Available when you open a data source and click the Summary tab.
- Attribute metadata. Available when you double-click an attribute from a data source's Attribute Details tab.
- Rule Set Association and Export Rule Sets windows.
Follow any one of the below procedure, depending on how you are adding the attribute:
- To add and edit a custom attribute column to a data source
-
- In the Add Data Source window, on the Configure tab, click Edit Columns. The Edit Columns window opens.
- Click the Add AttributeColumn icon (). The Add Attribute Column window opens.
- In Attribute Name, enter a name for the new attribute. The name must be unique. Using special characters in the name is not recommended.
- For Maximum Length, either keep the default of 100 or select another length. 100 is recommended.
- In Description, enter a brief description about the new attribute.
- In the Expression for Values field, build the expression to use to generate the attribute. For more information, see Creating and Editing Expressions.
- Click Save. The Add Attribute Column window closes and the new column is added to the Included Columns section. Change the order of columns as needed.
- Click Done. The Edit Columns window closes and the new attribute is added to the Data Preview. You may need to scroll through the view or change the column positions to see the new attribute column in the Data Preview.
- To edit the new attribute, click Edit Columns to open the Edit Attribute Column window. Select the new attribute in the Included Columns list and click the edit attribute icon (). The Edit Attribute Column window opens. Make updates as needed and click Save. Click Done to close the Edit Attribute Column window.
Note: After you finish adding the data source, you can only edit the new attribute's name and description. See Editing Attributes.
- To add and edit a new attribute column to a tab view
-
- Click the tab to which you want to add the custom column. For example, if you want to add a column to the data rows of a data source, open a data source and click the Data Rows tab.
- Click . The Choose Columns window opens.
- Click the Add Attribute Column icon (). The Add Attribute Column window opens.
- In Attribute Name, enter a name for the new column. The name must be unique and include no special characters.
- For Maximum Length, either keep the default of 100 or select another length. 100 is recommended.
- In Description, enter a brief description about the new column.
- In the Expression for Values field, build the expression to use to generate the column. For more information, see Creating and Editing Expressions.
- Click Save. The new column is added to the Included Columns section. Change the order of columns as needed.
- Click Done. The Add Attribute Column window closes and the new column is added to the tab view.
- To edit the new column at any time, click to open the Choose Columns window. Select the new column in the Included Columns list and click the edit attribute icon (). The Edit Attribute Column window opens. Make updates as needed and click Save. Click Done to close the Choose Columns window.