Please note: The algorithm descriptions in English have been automatically translated. Errors may have been introduced in this process. For the original descriptions, go to the Dutch version of the Algorithm Register.
Profile Re-enrolments A National Approach to Address Quality (LAA)
- Publication category
- Impactful algorithms
- Impact assessment
- Field not filled in.
- Status
- In use
General information
Theme
Begin date
Contact information
Link to publication website
Responsible use
Goal and impact
The purpose of this algorithm is to select potential high-risk addresses based on reasonable doubt and provide them to municipalities for further address searches. Based on these address searches, incorrect address data is corrected so that public bodies can perform their public duties based on accurate information.
Using the algorithm, you can prioritise addresses where there is a greater than 50 per cent chance that something is wrong in the registration and at the same time ensure that you do not visit citizens' doors unnecessarily. There is no automatic decision-making in this algorithm. There is always human intervention by the municipal official and the LAA partnership, which consists of signal providers, RvIG, the VNG and the NVVB.
If an enquiry into registration of persons at the address is going to take place, a home visit to the address is carried out which may involve personal privacy.
Based on its own address investigation, the municipality decides whether the address data in the BRP should be updated. A decision by the municipality can be appealed.
The profile does not use (personal) data that could lead to discrimination and does not select on nationality, place of birth or derivatives thereof.
Considerations
The accuracy of (address) data in the BRP is essential for the functioning of the Dutch government. A government that does not know or does not know how to find its citizens is unable to perform the tasks entrusted to it by its citizens. In many cases, it is a precondition for the correct implementation of legal regulations and for the effectiveness of government policies. For citizens, proper registration is important for their correspondence with the government and for claiming address-related government services.
The algorithm ensures efficient and most effective finding of addresses where the probability of incorrect registrations is high. Using the algorithm, you can prioritise addresses where the probability of incorrect registration is greater than 50 per cent while ensuring that you do not unnecessarily visit citizens' doors. If an address meets the algorithm's selection criteria, the address is shared with a municipality. Following the signal, the municipality may decide to launch an address investigation. This may include a home visit by the municipality. Selection of a high-risk address does not automatically lead to an adjustment of the address data in the BRP. It can only do so if the municipality sees reason to do so after conducting its own investigation. There is no automatic decision-making in this algorithm. There is always human intervention by the municipality's official and the LAA partnership, which consists of signal providers, RvIG, the VNG and the NVVB. The decision of a municipality to change the address data in the BRP can be objected to.
Human intervention
There is no automatic decision-making. Assessment and decisions are always done by humans. Algorithms are a tool and are not leading. Advice was sought from a representative representation of municipalities during the development of all LAA algorithms.
Risk management
Privacy is potentially compromised, due to the processing of personal data and home visits by municipalities.
Legal basis
Under the Basic Registration of Persons Act (BRP Act), the Minister of the Interior and Kingdom Relations may process personal data for the purpose of creating and applying algorithms. The Act and rules based on it specify what kind of personal data is involved. So the minister is not allowed to process all kinds of personal data just like that. The law and regulations set limits to this. For example, the minister may not process personal data relating to nationality, place of birth or data derived therefrom that results in discriminatory selection of persons. Nor may the minister process criminal data.
The General Data Protection Regulation (AVG) is also taken into account when creating algorithms.
RvIG carries out the task of processing personal data for LAA on behalf of the minister.
Links to legal bases
Operations
Data
The data from the Basic Registration of Persons (BRP) such as number of people registered at an address and the data from the Basic Registration of Addresses and Buildings (BAG) such as living area per address and the function and status of the address.
Regulations state which data we are allowed to use in the algorithm. Data that is not mentioned is therefore not allowed to be used. Nor do we use (personal) data that could lead to discrimination. We do not select on nationality, place of birth or derivatives thereof.
Technical design
The model uses a simple decision tree based on selection criteria. As input to the model, data from a copy of the BRP and the BAG are used, stored in a separate closed environment. The personal data in the BRP are partially pseudonymised along with the BAG identification code. The decision tree consists of a number of filters that are applied. In the first step, all addresses that have "cell function" according to the BAG are removed. Next, addresses are selected where at least 1 flaxing person is registered with a residential address that was 'Departed Unknown Whereabouts' (VOW) before registration and has been 'Departed Unknown Whereabouts' (VOW) at least 2 times in the past. At the time this person was registered, there was someone else registered at the address (regardless of whether this person has a residential or postal address). One of the 'Departed Unknown Whereabouts (VOW)' who is still registered there has been living there for less than the specified number of months. 2 or more household members are currently living at the address. A number of addresses are then excluded from this selection: addresses where only households (persons with a first-degree relationship registered in the BRP) are registered; addresses where the number of persons registered at the address falls below the minimum criterion; addresses that, according to the BAG, are not in use; addresses that, according to the BAG, have both the use function 'Health function' and 'Accommodation function'; addresses that, according to the BAG, have both the use function 'Health function' and 'Accommodation function'; addresses that, according to the BAG, have both the use function 'Meeting function' or 'Accommodation function' on which more than certain number of persons are registered.
The addresses remaining after this filtering are depseudonymised and delivered to the Information Exchange (IKP). In the IKP, addresses delivered to municipalities in the previous 365 days are removed. The IKP then checks, based on up-to-date information from the BRP and BAG, whether the addresses still meet the profile's selection criteria. The remaining addresses are further refined based on calculation rules developed from practical experience and past results. Addresses with a relatively low probability of incorrect registration are removed. Addresses with a relatively high probability of incorrect registration are delivered to the municipality. The municipality decides independently for each delivered address whether an address search is necessary or not. The municipality shares the results of its LAA work with the IKP. These results are analysed and used in consultation with the domain expert for further refinement and exclusion of addresses.
Similar algorithm descriptions
Transit address profile National Approach to Address Quality (LAA)
National Authority for Identity Data
The algorithm selects residential addresses on which there were many registrations in the past period: there may be doubts as to whether actual removals have taken place.Last change on 27th of May 2024, at 12:05 (CET) | Publication Standard 1.0- Publication category
- Impactful algorithms
- Impact assessment
- Field not filled in.
- Status
- In use
- The algorithm selects potential high-risk addresses on which several people with residential addresses are registered in the Basic Registration of Persons (BRP), who are not first-degree relatives of each other.Last change on 27th of May 2024, at 12:04 (CET) | Publication Standard 1.0
- Publication category
- Impactful algorithms
- Impact assessment
- Field not filled in.
- Status
- In use
- This algorithm is used within the BRP API Occupancy to provide the different occupancy situations at an address over time. The query can be asked with a given reference date or with a given time period.Last change on 21st of October 2024, at 10:44 (CET) | Publication Standard 1.0
- Publication category
- Impactful algorithms
- Impact assessment
- DPIA
- Status
- In use
- This algorithm is used within the BRP API Persons to provide parental authority over a minor.Last change on 21st of October 2024, at 13:59 (CET) | Publication Standard 1.0
- Publication category
- Impactful algorithms
- Impact assessment
- DPIA
- Status
- In use
- The search algorithm used by the BRP Link Point to determine whether a person logging in with an eIDAS resource already appears in the Basic Registration of Persons (BRP).Last change on 15th of October 2024, at 8:31 (CET) | Publication Standard 1.0
- Publication category
- Other algorithms
- Impact assessment
- DPIA
- Status
- In use