View grammars found
From the Data Analysis tab () of the content detail panel, you can view a list of grammar classes and rules found in data items that have been analyzed and have had grammars extracted.
For unstructured data items, you can view the grammar classes, rules, and values identified by the rules. For structured data items, you can view the grammar rules and values identified by the rules and the name of the columns in which a grammar values have been identified.
The number of values next to the grammar rule name is the total number of values identified in the document and any attachments. The percentage next to the grammar value represents the calculation of the confidence level that the identified value is sensitive content.
To view the identified grammar rules and values, select a document or table from the content list in either the List view () or Detailed view (
). In the content detail panel, click the Data Analysis tab (
).
NOTE: If content has not been stored, you can see the grammar rules and values but you cannot select a rule or value to highlight it in the content panel.
-
For unstructured content, expand grammar classes and rules to view the grammar values found in the document.
For structured content, expand the column names in the Columns with Sensitive Data section and then expand the grammar rules to view the grammar values found in the table. Columns in which no grammars were found are listed in the Other Columns section.
-
Click one or more rows for grammar rules or values. The background color for the clicked row changes to blue to indicate that the rule or value is "selected". If the content has been stored, the associated grammar values are highlighted in the content view panel.
-
If you click a grammar rule, all values for all grammar are selected in the grammar rules list and are highlighted in the content view panel.
TIP: If your content list view is a result of a search that included a keyword search, the highlight color for grammar rule identification is a different color than the highlight color used to identify found keywords in collected documents.
-
If you click a grammar value, the value is highlighted in both in the grammar rules list and is highlighted in the content view panel. If more than one instance of the value is present in the content, go to the next value using the navigation arrows at the bottom of the panel (
1 of 2
).
-
A red box around a highlighted value (word or phrase) in the content view indicates the grammar value in focus.
-
-
If you have multiple grammar values or rules and want to change the focus of a selected value, hover over a selected grammar value and then click the view icon (
). The focus in the content view panel changes to the selected value. If more than one instance of the value is present in the content, go to the next value using the navigation arrows at the bottom of the panel (
1 of 2
).
-
To deselect a grammar rule or value, click the desired rule or value. The blue background is removed to indicate that the rule or value is no longer selected.
TIP: If a value remains highlighted in the content view panel after you deselected it in the grammar rules list, the value is still selected for another grammar rule.
False positives
When reviewing data items, you may find that some grammar values identified for a particular rule display in strikethrough text (sample). For these values, a Manage reviewer determined that the identified value was not sensitive in context. For example, the last name "White" may have been identified as a value matching the grammar rule "Racial Ethnic Origin" within the "Other Personal Data" grammar class. The reviewer marked the grammar value as a false positive because, in context, the last name is not sensitive for the rule "Racial Ethnic Origin". The reviewer may have also marked an entire grammar rule as a false positive if all identified grammar values for the rule were not correct in context.
Not sensitive table columns
Similar identified grammar values marked as false positives, columns in structured data tables may display in strikethrough text (sample). For these columns, a Manage reviewer determined that the entire column is not sensitive in context. For example, the database table you are viewing includes customer account numbers. The grammar analysis of this table identified all values in the "Customer #" column as bank account numbers because the customer account numbers happen to have the same pattern as Australian bank account numbers. When reviewing this table, the reviewer set this entire column as "not sensitive" because the information in the column is not sensitive for the identified rule. All grammar rules and values in the column marked as not sensitive are marked as false positive and also display in strikethrough text.