Skip to main content

Set Your Preferences

You can configure default preferences that will be automatically applied to every anonymization run. These settings help tailor the format and output of the generated documents to your specific needs.

Setting Preferences​

Via the Web Client​

  1. Navigate to Account → User Preferences in the web interface
  2. Configure your default preferences
  3. Save your changes

These preferences will apply to all future runs initiated through the web client.

Via the Python Client​

Set preferences when creating a runner using manager.create_runner():

from avatars.client import Client
from avatars.models import DataRecipient, DataType, DataSubject, SensitivityLevel, ReportLanguage

client = Client()
manager = client.managers.get("my-manager")

runner = manager.create_runner(
    max_distribution_plots=20,
    pia_data_recipient=DataRecipient.INTERNAL,
    pia_data_type=DataType.HEALTH,
    pia_data_subject=DataSubject.PATIENTS,
    pia_sensitivity_level=SensitivityLevel.HIGH,
    report_language=ReportLanguage.EN
)

Available Preferences​

max_distribution_plots​

Type: int | None
Default: None

Maximum number of distribution plots to include in the quality report. Set to limit report size when working with datasets containing many columns.

report_language​

Type: ReportLanguage
Default: ReportLanguage.EN

Language for the generated quality report.

Available values:

  • ReportLanguage.EN - English
  • ReportLanguage.FR - French

Privacy Impact Assessment (PIA) Metadata​

The following preferences help document the context of your data processing for privacy assessment and compliance purposes. These values are included in the anonymization report.

pia_sensitivity_level​

Type: SensitivityLevel
Default: SensitivityLevel.UNDEFINED

Evaluation of the sensitivity level of the personal data being processed, based on the nature of the data and potential risks to data subjects.

Available values:

  • SensitivityLevel.VERY_HIGH - Highly sensitive data with severe re-identification risks
  • SensitivityLevel.HIGH - Sensitive personal data (e.g., GDPR Art. 9 special categories)
  • SensitivityLevel.MEDIUM - Personal data with moderate sensitivity
  • SensitivityLevel.LOW - Low sensitivity demographic data
  • SensitivityLevel.VERY_LOW - Minimal sensitivity data
  • SensitivityLevel.NEGLIGIBLE - Non-sensitive data
  • SensitivityLevel.UNDEFINED - Sensitivity not yet assessed

The chosen value will directly influence the content of the automatically generated impact analysis alongside the anonymization, and will help you document the associated risks and mitigation measures for your project.

tip

Sensitive personal data under GDPR Art. 9 (health, racial/ethnic origin, political opinions, religious beliefs, genetic data, biometric data, etc.) typically require VERY_HIGH or HIGH sensitivity levels.

pia_data_recipient​

Type: DataRecipient
Default: DataRecipient.UNKNOWN

Categories of recipients who will have access to the anonymized data.

Available values:

  • DataRecipient.INTERNAL - Internal stakeholders within the same organization
  • DataRecipient.CONTRACTUAL_THIRDPARTY - Third parties with contractual relationship (research partners, vendors)
  • DataRecipient.TRUSTED_THIRDPARTY - Trusted third parties under confidentiality agreements
  • DataRecipient.OPENDATA - General public through open data repositories
  • DataRecipient.OUTSIDE_EU - Entities outside the European Union
  • DataRecipient.UNKNOWN - Recipients not yet identified

The chosen value will allow generating a targeted paragraph in the privacy impact analysis report.

pia_data_subject​

Type: DataSubject
Default: DataSubject.UNKNOWN

Categories of individuals whose personal data are being processed.

Available values:

  • DataSubject.PATIENTS - Patients in healthcare or clinical research contexts
  • DataSubject.EMPLOYEES - Employees, job applicants, or contractors
  • DataSubject.CLIENTS - Clients, customers, or insured persons
  • DataSubject.USERS - Users of digital, public, or mobility services
  • DataSubject.STUDENTS - Students in educational institutions
  • DataSubject.UNKNOWN - Data subjects not yet categorized

The chosen value will allow generating a targeted paragraph in the privacy impact analysis report.

pia_data_type​

Type: DataType
Default: DataType.UNKNOWN

Categories of personal data being processed, based on the sector and context.

Available values:

  • DataType.HEALTH - Health-related datasets with clinical and demographic variables
  • DataType.HR - Human resources data (employment history, remuneration, performance)
  • DataType.INSURANCE - Insurance policyholder and claims data
  • DataType.FINANCE - Financial service data (transactions, accounts, credit ratings)
  • DataType.MOBILITY - Transport and mobility service data (geolocation, travel routes)
  • DataType.EDUCATION - Educational data (academic performance, attendance)
  • DataType.UNKNOWN - Data type not yet determined

The chosen value will allow generating a targeted paragraph in the privacy impact analysis report.

Example: Complete Configuration​

runner = manager.create_runner(
    max_distribution_plots=15,
    pia_data_recipient=DataRecipient.CONTRACTUAL_THIRDPARTY,
    pia_data_type=DataType.HEALTH,
    pia_data_subject=DataSubject.PATIENTS,
    pia_sensitivity_level=SensitivityLevel.VERY_HIGH,
    report_language=ReportLanguage.FR
)

This configuration:

  • Limits distribution plots to 15
  • Indicates data shared with contractual research partners
  • Specifies health data about patients
  • Marks data as very high sensitivity
  • Generates reports in French