LogoLogo
2024.2
  • Immuta Documentation - 2024.2
  • What is Immuta?
  • Self-Managed Deployment
    • Getting Started
    • Deployment Requirements
    • Install
      • Managed Public Cloud
      • Red Hat OpenShift
      • Generic Installation
      • Immuta in an Air-Gapped Environment
      • Deploy Immuta without Elasticsearch
    • Configure
      • Ingress Configuration
      • Cosign Verification
      • TLS Configuration
      • Immuta in Production
      • External Cache Configuration
      • Rotating Credentials
      • Enabling Legacy Query Engine and Fingerprint
    • Upgrade
      • Upgrade Immuta
      • Upgrade to Immuta 2024.2 LTS
    • Disaster Recovery
    • Troubleshooting
    • Conventions
    • Release Notes
  • Data and Integrations
    • Immuta Integrations
    • Snowflake
      • Getting Started
      • How-to Guides
        • Configure a Snowflake Integration
        • Snowflake Table Grants Migration
        • Edit or Remove Your Snowflake Integration
        • Integration Settings
          • Enable Snowflake Table Grants
          • Use Snowflake Data Sharing with Immuta
          • Configure Snowflake Lineage Tag Propagation
          • Enable Snowflake Low Row Access Policy Mode
            • Upgrade Snowflake Low Row Access Policy Mode
      • Reference Guides
        • Snowflake Integration
        • Snowflake Data Sharing
        • Snowflake Lineage Tag Propagation
        • Snowflake Low Row Access Policy Mode
        • Snowflake Table Grants
        • Warehouse Sizing Recommendations
      • Phased Snowflake Onboarding Concept Guide
    • Databricks Unity Catalog
      • Getting Started
      • How-to Guides
        • Configure a Databricks Unity Catalog Integration
        • Migrate to Unity Catalog
      • Databricks Unity Catalog Integration Reference Guide
    • Databricks Spark
      • How-to Guides
        • Configuration
          • Simplified Databricks Configuration
          • Manual Databricks Configuration
          • Manually Update Your Databricks Cluster
          • Install a Trusted Library
        • DBFS Access
        • Limited Enforcement in Databricks
        • Hide the Immuta Database in Databricks
        • Run spark-submit Jobs on Databricks
        • Configure Project UDFs Cache Settings
        • External Metastores
      • Reference Guides
        • Databricks Spark Integration
        • Databricks Spark Pre-Configuration Details
        • Configuration Settings
          • Cluster Policies
            • Python & SQL
            • Python & SQL & R
            • Python & SQL & R with Library Support
            • Scala
            • Sparklyr
          • Environment Variables
          • Ephemeral Overrides
          • Py4j Security Error
          • Scala Cluster Security Details
          • Databricks Security Configuration for Performance
        • Databricks Change Data Feed
        • Databricks Libraries Introduction
        • Delta Lake API
        • Spark Direct File Reads
        • Databricks Metastore Magic
    • Starburst (Trino)
      • Getting Started
      • How-to Guides
        • Configure Starburst (Trino) Integration
        • Customize Read and Write Access Policies for Starburst (Trino)
      • Starburst (Trino) Integration Reference Guide
    • Redshift
      • Getting Started
      • How-to Guides
        • Configure Redshift Integration
        • Configure Redshift Spectrum
      • Reference Guides
        • Redshift Integration
        • Redshift Pre-Configuration Details
    • Azure Synapse Analytics
      • Getting Started
      • Configure Azure Synapse Analytics Integration
      • Reference Guides
        • Azure Synapse Analytics Integration
        • Azure Synapse Analytics Pre-Configuration Details
    • Amazon S3
    • Google BigQuery
    • Legacy Integrations
      • Securing Hive and Impala Without Sentry
      • Enabling ImmutaGroupsMapping
    • Registering Metadata
      • Data Sources in Immuta
      • Register Data Sources
        • Create a Data Source
        • Create an Amazon S3 Data Source
        • Create a Google BigQuery Data Source
        • Bulk Create Snowflake Data Sources
      • Data Source Settings
        • How-to Guides
          • Manage Data Sources and Data Source Settings
          • Manage Data Source Members
          • Manage Access Requests and Tasks
          • Manage Data Dictionary Descriptions
          • Disable Immuta from Sampling Raw Data
        • Data Source Health Checks Reference Guide
      • Schema Monitoring
        • How-to Guides
          • Run Schema Monitoring and Column Detection Jobs
          • Manage Schema Monitoring
        • Reference Guides
          • Schema Monitoring
          • Schema Projects
        • Why Use Schema Monitoring?
    • Catalogs
      • Getting Started with External Catalogs
      • Configure an External Catalog
      • Reference Guides
        • External Catalogs
        • Custom REST Catalogs
          • Custom REST Catalog Interface Endpoints
    • Tags
      • How-to Guides
        • Create and Manage Tags
        • Add Tags to Data Sources and Projects
      • Tags Reference Guide
  • People
    • Getting Started
    • Identity Managers (IAMs)
      • How-to Guides
        • Microsoft Entra ID
        • Okta LDAP Interface
        • Okta and OpenID Connect
        • Integrate Okta SAML SCIM with Immuta
        • OneLogin with OpenID
        • Configure SAML IAM Protocol
      • Reference Guides
        • Identity Managers
        • SAML Single Logout
        • SAML Protocol Configuration Options
    • Immuta Users
      • How-to Guides
        • Managing Personas and Permissions
        • Manage Attributes and Groups
        • User Impersonation
        • External User ID Mapping
        • External User Info Endpoint
      • Reference Guides
        • Attributes and Groups in Immuta
        • Permissions and Personas
  • Discover Your Data
    • Getting Started
    • Introduction
    • Architecture
    • Data Discovery
      • How-to Guides
        • Enable Sensitive Data Discovery (SDD)
        • Manage Identification Frameworks
        • Manage Patterns
        • Manage Rules
        • Manage SDD on Data Sources
        • Manage Global SDD Settings
        • Migrate From Legacy to Native SDD
      • Reference Guides
        • How Competitive Pattern Analysis Works
        • Built-in Pattern Reference
        • Built-in Discovered Tags Reference
    • Data Classification
      • How-to Guides
        • Activate Classification Frameworks
        • Adjust Identification and Classification Framework Tags
        • How to Use a Built-In Classification Framework with Your Own Tags
      • Built-in Classification Frameworks Reference Guide
  • Detect Your Activity
    • Getting Started
      • Monitor and Secure Sensitive Data Platform Query Activity
        • User Identity Best Practices
        • Integration Architecture
        • Snowflake Roles Best Practices
        • Register Data Sources
        • Automate Entity and Sensitivity Discovery
        • Detect with Discover: Onboarding Guide
        • Using Immuta Detect
      • General Immuta Configuration
        • User Identity Best Practices
        • Integration Architecture
        • Databricks Roles Best Practices
        • Register Data Sources
    • Introduction
    • Audit
      • How-to Guides
        • Export Audit Logs to S3
        • Export Audit Logs to ADLS
        • Run Governance Reports
      • Reference Guides
        • Universal Audit Model (UAM)
        • Snowflake Query Audit Logs
        • Databricks Unity Catalog Audit Logs
        • Databricks Query Audit Logs
        • Starburst (Trino) Query Audit Logs
        • UAM Schema
        • Audit Export CLI
        • Governance Report Types
      • Deprecated Audit Guides
        • Legacy to UAM Migration
        • Download Audit Logs
        • System Audit Logs
    • Detection
      • Use the Detect Dashboards
      • Reference Guides
        • Detect
        • Detect Dashboards
        • Unknown Users in Audit Logs
    • Monitors
      • Manage Monitors and Observations
      • Detect Monitors Reference Guide
  • Secure Your Data
    • Getting Started with Secure
      • Automate Data Access Control Decisions
        • The Two Paths: Orchestrated RBAC and ABAC
        • Managing User Metadata
        • Managing Data Metadata
        • Author Policy
        • Test and Deploy Policy
      • Compliantly Open More Sensitive Data for ML and Analytics
        • Managing User Metadata
        • Managing Data Metadata
        • Author Policy
      • Federated Governance for Data Mesh and Self-Serve Data Access
        • Defining Domains
        • Managing Data Products
        • Managing Data Metadata
        • Apply Federated Governance
        • Discover and Subscribe to Data Products
    • Introduction
      • Scalability and Evolvability
      • Understandability
      • Distributed Stewardship
      • Consistency
      • Availability of Data
    • Authoring Policies in Secure
      • Authoring Policies at Scale
      • Data Engineering with Limited Policy Downtime
      • Subscription Policies
        • How-to Guides
          • Author a Subscription Policy
          • Author an ABAC Subscription Policy
          • Subscription Policies Advanced DSL Guide
          • Author a Restricted Subscription Policy
          • Clone, Activate, or Stage a Global Policy
        • Reference Guides
          • Subscription Policies
          • Subscription Policy Access Types
          • Advanced Use of Special Functions
      • Data Policies
        • Overview
        • How-to Guides
          • Author a Masking Data Policy
          • Author a Minimization Policy
          • Author a Purpose-Based Restriction Policy
          • Author a Restricted Data Policy
          • Author a Row-Level Policy
          • Author a Time-Based Restriction Policy
          • Certifications Exemptions and Diffs
          • External Masking Interface
        • Reference Guides
          • Data Policy Types
          • Masking Policies
          • Row-Level Policies
          • Custom WHERE Clause Functions
          • Data Policy Conflicts and Fallback
          • Custom Data Policy Certifications
          • Orchestrated Masking Policies
    • Domains
      • Getting Started with Domains
      • Domains Reference Guide
    • Projects and Purpose-Based Access Control
      • Projects and Purpose Controls
        • Getting Started
        • How-to Guides
          • Create a Project
          • Create and Manage Purposes
          • Adjust a Policy
          • Project Management
            • Manage Projects and Project Settings
            • Manage Project Data Sources
            • Manage Project Members
        • Reference Guides
          • Projects and Purposes
          • Policy Adjustments
        • Why Use Purposes?
      • Equalized Access
        • Manage Project Equalization
        • Project Equalization Reference Guide
        • Why Use Project Equalization?
      • Masked Joins
        • Enable Masked Joins
        • Why Use Masked Joins?
      • Writing to Projects
        • How-to Guides
          • Create and Manage Snowflake Project Workspaces
          • Create and Manage Databricks Project Workspaces
          • Write Data to the Workspace
        • Reference Guides
          • Project Workspaces
          • Project UDFs (Databricks)
    • Data Consumers
      • Subscribe to a Data Source
      • Query Data
        • Querying Snowflake Data
        • Querying Databricks Data
        • Querying Databricks SQL Data
        • Querying Starburst (Trino) Data
        • Querying Redshift Data
        • Querying Azure Synapse Analytics Data
      • Subscribe to Projects
  • Application Settings
    • How-to Guides
      • App Settings
      • BI Tools
        • BI Tool Configuration Recommendations
        • Power BI Configuration Example
        • Tableau Configuration Example
      • Add a License Key
      • Add ODBC Drivers
      • Manage Encryption Keys
      • System Status Bundle
    • Reference Guides
      • Data Processing, Encryption, and Masking Practices
      • Metadata Ingestion
  • Releases
    • Immuta v2024.2 Release Notes
    • Immuta Release Lifecycle
    • Immuta LTS Changelog
    • Immuta Support Matrix Overview
    • Immuta CLI Release Notes
    • Immuta Image Digests
    • Preview Features
      • Features in Preview
    • Deprecations
  • Developer Guides
    • The Immuta CLI
      • Install and Configure the Immuta CLI
      • Manage Your Immuta Tenant
      • Manage Data Sources
      • Manage Sensitive Data Discovery
        • Manage Sensitive Data Discovery Rules
        • Manage Identification Frameworks
        • Run Sensitive Data Discovery on Data Sources
      • Manage Policies
      • Manage Projects
      • Manage Purposes
    • The Immuta API
      • Integrations API
        • Getting Started
        • How-to Guides
          • Configure an Amazon S3 Integration
          • Configure an Azure Synapse Analytics Integration
          • Configure a Databricks Unity Catalog Integration
          • Configure a Google BigQuery Integration
          • Configure a Redshift Integration
          • Configure a Snowflake Integration
          • Configure a Starburst (Trino) Integration
        • Reference Guides
          • Integrations API Endpoints
          • Integration Configuration Payload
          • Response Schema
          • HTTP Status Codes and Error Messages
      • Immuta V2 API
        • Data Source Payload Attribute Details
        • Data Source Request Payload Examples
        • Create Policies API Examples
        • Create Projects API Examples
        • Create Purposes API Examples
      • Immuta V1 API
        • Authenticate with the API
        • Configure Your Instance of Immuta
          • Get Fingerprint Status
          • Get Job Status
          • Manage Frameworks
          • Manage IAMs
          • Manage Licenses
          • Manage Notifications
          • Manage Sensitive Data Discovery (SDD)
          • Manage Tags
          • Manage Webhooks
          • Search Filters
        • Connect Your Data
          • Create and Manage an Amazon S3 Data Source
          • Create an Azure Synapse Analytics Data Source
          • Create an Azure Blob Storage Data Source
          • Create a Databricks Data Source
          • Create a Presto Data Source
          • Create a Redshift Data Source
          • Create a Snowflake Data Source
          • Create a Starburst (Trino) Data Source
          • Manage the Data Dictionary
        • Manage Data Access
          • Manage Access Requests
          • Manage Data and Subscription Policies
          • Manage Domains
          • Manage Write Policies
            • Write Policies Payloads and Response Schema Reference Guide
          • Policy Handler Objects
          • Search Audit Logs
          • Search Connection Strings
          • Search for Organizations
          • Search Schemas
        • Subscribe to and Manage Data Sources
        • Manage Projects and Purposes
          • Manage Projects
          • Manage Purposes
        • Generate Governance Reports
Powered by GitBook

Other versions

  • SaaS
  • 2024.3

Copyright © 2014-2024 Immuta Inc. All rights reserved.

On this page
  • Sampling
  • Sampling considerations
  • Qualifying
  • Scoring
  • Example
  • Important notes

Was this helpful?

Export as PDF
  1. Discover Your Data
  2. Data Discovery
  3. Reference Guides

How Competitive Pattern Analysis Works

PreviousReference GuidesNextBuilt-in Pattern Reference

Last updated 6 months ago

Was this helpful?

Of sensitive data discovery's , regex and dictionary are competitive. This means that when assessing your data, if multiple patterns could match, only one of the competitive patterns will be chosen and tag the data. To better understand how Immuta executes this competition, read further.

Discover employs a three-phased competitive pattern analysis approach for sensitive data discovery (SDD):

  1. : No data is moved, and Immuta checks the patterns against a sample of data from the table.

  2. : Patterns that have less than a 90% match are filtered out.

  3. : The remaining patterns are compared with one another to find the most specific pattern that qualifies and matches the sample.

In the end, competitive pattern analysis aims to find a single pattern for each column that best describes the data format.

Sampling

In the sampling process, no database contents are transmitted to Immuta; instead, Immuta receives only the column-wise hit rate (the number of times the pattern has matched a value in the column) information for each active pattern. To do this, Discover instructs a remote database to measure column-wise hit rate information for all active patterns over a row sample.

The sample size is decided based on the number of patterns and the data size, when available. In the most simplified case, the requested number of sampled rows depends only on the number of regex and dictionary patterns being run in the framework, not the data size. The sample size dependence on the number of patterns is weak and will not exceed 13,000 rows.

Number of patterns
Sample size

5

7369 rows

50

9211 rows

500

11053 rows

5000

12895 rows

Sampling considerations

In practice, the number of sampled values for each column may be less than the requested number of rows. This happens when the target table has less than the requested number of rows, when many of the column values are null, or because of technology-specific limitations.

  • Snowflake and Starburst (Trino): Discover implements native table sampling by row count.

  • Databricks and Redshift: Due to technology limitations and the inability to predict the size of the table, Discover implements a best-effort sampling strategy comprising a flat 10% row sample capped at the first 10,000 sampled rows. In particular, under-sampling may occur on tables with less than 100,000 rows. Moreover, the resulting sample is biased towards earlier records.

  • All platforms: Sampling from views can have significantly slower performance that varies by the performance of the query that defines the view.

Qualifying

During the qualification phase, patterns that do not agree with the data are disqualified. A pattern agrees with the data if the hit rate on the remote sample exceeds the predefined threshold. This threshold is 90% match for most built-in patterns; however, two built-in patterns have lower threshold . The 90% threshold is standard for all custom patterns as well to ensure the pattern matches the data within the column and avoid false positives. If no patterns qualify, then no pattern is assessed for scoring and the column is not tagged.

Scoring

During the scoring phase, a machine inference is carried out among all qualified patterns, combining pattern-derived complexity information with hit rate information to determine which pattern best describes the sample data. This process prefers the more restrictive of two competing patterns since the ability to satisfy the more difficult-to-satisfy pattern itself serves as evidence that it is more likely. This phase ends by returning a single most likely pattern per the inference process.

Example

Here are a set of regex patterns and a sample of data:

Patterns:

  1. [a-zA-Z0-9]{3} - This pattern will match 3 character strings with the characters a-z, lowercase or uppercase, or digits 0-9.

  2. [a-c]{3} - This pattern will match 3 character strings with the characters a-c, lowercase.

  3. (a|b|d){3} - This pattern will match 3 character strings with the characters a, b, or d, lowercase.

Sample data
Matches Pattern 1
Matches Pattern 2
Matches Pattern 3

dad

Yes

Yes

baa

Yes

Yes

add

Yes

Yes

add

Yes

Yes

cab

Yes

Yes

bad

Yes

Yes

aba

Yes

Yes

baa

Yes

Yes

dad

Yes

Yes

baa

Yes

Yes

When qualifying the patterns, Pattern 1 and Pattern 3 both match 90% or more of the data. Pattern 2 does not, and is disqualified.

Then the qualified patterns are scored. Here, Pattern 1, despite matching 100% of the data, is unspecific and could match over 200,000 values. On the other hand, Pattern 3 matches just at 90% but is very specific with only 27 available values.

Therefore, with the specificity taken into account, Pattern 3 would be the match for this column, and its tags would be applied to the data source in Immuta.

Important notes

  • Dictionaries are considered patterns by Immuta and are part of the competitive process, while column-name regex patterns are not.

  • Scoring ties are rare but can occur if the same pattern is specified more than once (even in different forms). Scoring ties are inconclusive, and the scoring phase will not return a pattern in the case of a tie.

  • Pattern complexity analysis is sensitive to the total number of strings a pattern accepts or, equivalently for dictionaries, the number of entries. Therefore, patterns that accept much more than is necessary to describe the intended column data format may perform more poorly in the competitive analysis because they are easier to satisfy.

❌
❌
❌
❌
❌
❌
❌
❌
❌
❌
Sampling
Qualifying
Scoring
three pattern options