Anonymisation tool

Municipality of Noord-Beveland

The algorithm recognises and anonymises personal and, in specific cases, confidential business data in documents before they are published.

Last change on 5th of March 2024, at 14:34 (CET) | Publication Standard 1.0

• Publication category: Impactful algorithms
• Impact assessment: DPIA
• Status: In use

General information

Theme

Organisation and business operations

Begin date

12-2023

Contact information

[e-mail]

Responsible use

Goal and impact

The purpose of the algorithm is to make the municipality's operations faster, more efficient and cheaper. We think the impact on citizens and businesses is not very big, because it does not change the outcome. The work of anonymising documents is mainly faster.

Considerations

One advantage of deploying this algorithm is that personal and business data can be masked much faster and better. The alternative is masking everything manually. This costs an employee much more time.

Human intervention

Employee must assess and monitor final result

Risk management

Employee must assess and monitor final result

Links to legal bases

• AVG: https://wetten.overheid.nl/BWBR0011468
• UAVG: https://wetten.overheid.nl/BWBR0040940
• WOO: https://wetten.overheid.nl/BWBR0045754
• Wep: https://wetten.overheid.nl/BWBR0043961

Elaboration on impact assessments

DPIA conducted

Impact assessment

Data Protection Impact Assessment (DPIA)

Operations

Data

This depends on the document being anonymised. Examples include personal data such as e-mail addresses, phone numbers, bank account numbers, address details and signatures. And based on the Open Government Act (Woo), it can also include data beyond personal data. These grounds for exception are listed in the Woo.

Technical design

Smart features, such as set rules or templates, make it possible to anonymise per document or as a bulk. In this way, the method and degree of anonymisation of commonly used (standard) documents can also be set. The software then uses pattern recognition and Natural Language Processing to search for names, addresses, dates of birth, specific set words, signatures or regular expressions (such as e-mail, IBAN, BSN). The DataMask software recognises these and makes suggestions to mask or anonymise them fully automatically.

External provider

DataMask B.V.