Cisco Secure Access Help

PDF

Cisco Secure Access Help

Create a Data Classification

Updated: November 8, 2026

Want to summarize with AI?

Log in

Describes Create a Data Classification in Cisco Secure Access. You can create a data classification to help you monitor content with specific characteristics.

You can create a data classification to help you monitor content with specific characteristics. Custom data classifications can be used in real time rules, SaaS API rules and discovery scans.

The building blocks for data classifications are data identifiers. The data identifiers that you choose for a data classification determine the type of data for which rules using that data classification will scan.

Using the inclusion and exclusion options on the Data Classification page, you can fine-tune your classification to be more precise and reduce false positives. You can exclude specific terms and regular expression (regex) patterns by creating a custom identifier in the exclusion area, or by excluding a pre-existing built-in identifier.

The exclusion applies only to the content that's been matched, not to every document that meets the exclusion criteria.

For example, consider a data classification that targets the built-in identifier Health Condition and Person Name (US). You want to block instances of "John Smith cancer" while allowing instances of "John Smith cancer fundraising." To achieve this, you can craft a custom data identifier for "cancer fundraising" and set up an exclusion for this identifier within your data classification. As a result, matches for the Health Condition and Person Name (US) identifier will be flagged, except when the phrase "cancer fundraising" is present.

The system compares data identifiers selected for exclusion against both the keywords and the proximity terms of included data identifiers. If the Data Loss Prevention Report reveals that a particular rule or identifier is generating false positives, consider using terms and identifiers exclusion to remedy the situation.

Note
If you select a data identifier for both inclusion and exclusion, exclusion will take precedence.

The system offers two types of built-in data identifiers you can choose from:

Built-In Identifiers

These identify data using pattern matching and dictionary lookups. The descriptions shown in the GUI provide details about the type of data they match. For more information, see Built-In Data Identifiers.

Machine Learning Identifiers

These identify data based on AI analysis of example documents. For example, the identifier for Patent Files has been trained to recognize documents that are likely patent applications. For more information, see Built-In Data Identifiers.

The system offers three types of data identifiers you can create yourself applying different methods of data analysis:

Custom Identifiers

You can create custom identifiers to match specific terms and pattern expressions of your choosing. See Create a Custom Identifier .

Exact Data Match Identifiers

Exact Data Match Identifiers use fingerprinting to identify data in structured documents that match criteria you define. (See Create an Exact Data Match Identifier for more information.)

Indexed Document Match Identifiers

Indexed Document Match Identifiers use fingerprinting to identify data in unstructured documents that match criteria you define. See Create an Indexed Document Match Identifier for more information.

To delete or edit a data classification, see Delete or Edit a Classification.

Procedure for Creating a Data Classification

Before you begin

Procedure

  1. Navigate to Secure > Settings > Data Loss Prevention Classifications > Data Classification and click Add.


    Data Classification page displaying options for Data Classifications, Exact Data Matches, and Indexed Document Matches

  2. Give your classification a meaningful name and description.

  3. Under Identifier Group 1, select a Boolean operator to separate the identifiers included in the classification.

    • OR—At least one of the data identifiers included must match during rule evaluation.

    • AND—All of the data identifiers included must match during rule evaluation.


    Add New Data Classification page displaying options for Include Data Identifiers and Built-in Data Identifiers

  4. Click Save.

What to do next

Your new classification is listed on the Data Classification page, and will be available to you when you Add a Real Time Rule to the Data Loss Prevention Policy, Add a SaaS API Rule to the Data Loss Prevention Policy, or initiate a new Discovery Scan.

For multilingual OCR use cases, you can create separate classifications for each language or combine multiple language-specific identifiers into a single classification, depending on how you want the policy to be applied.