Ir para o conteúdo

Datavalid API Docs

What is Datavalid?

Datavalid is an intelligence solution that helps with the registration qualification of individuals and companies.

The Datavalid solution is widely used to streamline customer onboarding processes and minimize fraud rates. By verifying the data provided by the customer during registration, you can ensure the quality of the data in your database.

Datavalid can also be utilized to automate access control in physical environments or applications.

Types of validation

Cadastral and Biographical Validation:

  • Validating registration information involves checking the accuracy and completeness of various details such as the name, date of birth, address, affiliation, and document status, among others. The biographical data can refer to information about an individual (PF) or a legal entity (CNPJ). Find out more in fields available for validation.

Biometric validation:

  • Validation of drivers and citizens with a driving license (CNH) through facial recognition is being implemented.
  • Fingerprint validation is required for drivers and citizens who possess a CNH (driver's license) in order to verify their identity.

How does it work?

The interested party sends the data or image that he wants to validate through an API (application programming interface). For this, it is necessary to implement the call code as in the documentation.

Calling

The service contractor should collect the data it needs to validate through its data collection interface, like an app. After collecting the data, the contractor needs to send it for validation by making an API (application programming interface) call. DataValid compares the data received against the information available in the government database (such as CNH, CPF, CNPJ) and provides the contractor with the result of the comparison.

Returns

For biometric data, such as face images or fingerprints, the results will include the availability of the data in the database, the percentage of similarity in the biometric comparison (1:1), and the probability rating, which can be categorized as very low, low, high, or very high. These returns apply to both face image and fingerprints validation. Learn more about probability ranges.

  • For objective biographical data, the return will be true/false. The objective fields are those that have specific rules for completion, such as date of birth or driver's license category, for example.

  • For descriptive biographical data, the return of the comparison will be the percentage of similarity and true/false. Descriptive fields are those that have no fill rule, such as the name field, for example.

See below the example of field validation “nome” (name) and “data_nascimento” (date of birth) registered on the sandbox as “Manuela Elisa da Mota” and “04/06/1975”, respectively.

Entry: 

"nome": "Manuela Elisa da Mota" 
"data_nascimento": "1975-06-04"

Return: 

"nome": true, 
"nome_similaridade": 1 
"data_nascimento": true

For more technical information about Datavalid, see documentation sections.

Characteristics

  • There is no volume limit for requests.
  • Regarding facial validation, Datavalid deals exclusively with the facial comparison of the face image sent by the client against the face image recorded in the government base to the informed CPF.
  • Datavalid utilizes the Serpro biometric validation engine, which has a facial validation accuracy of 99.9%.
  • The validation of data by biometric comparison is a probabilistic and non-deterministic resource, and it is up to the contractor to decide on the use of the results returned by the service.
  • Datavalid only processes the data sent by the contractor.
  • The contractor needs to connect its data collection systems to Datavalid through Web API REST and JSON calls.
  • It is possible to limit the usage of APIs by implementing a maximum request limit for their operation. Learn more by accessing the Client Area / API Consumption Restriction option in the Help Center.

What Datavalid does not do

  • Datavalid does not check if the image of the face sent by the customer suffered any kind of manipulation, as virtual camera capture, artificial intelligence, as mix between faces, Deepfake, image injection, among other techniques. It is noteworthy that the protection against identity theft is the responsibility of the data collection process (image of the face), in order to ensure that it collected an image that is a legit real person image before submitting it to the Datavalid validation API;
  • Datavalid does not return the database data used in validation;
  • Datavalid does not determine customers' business rules.
  • Datavalid does not deliver a frontend data collection system.

Datavalid demonstration

Here you can find information about the free test and demo features offered by Datavalid. Tutorial how to use the Datavalid Demo.

How to sign up

If you want to integrate Datavalid to your system, please send a message to internacional@serpro.gov.br

Documentation

This page offers technical information about Datavalid. To find the desired information, use the search field, navigate the menu on the left, or start with the items below.