We’re happy to announce the latest we’ve brought to activities and hope they improve the experience for you & for the business users, validating documents! 
The new release comes with the following enhancements:
Validate Document Classification in the new cross-platform Activities
The newest DocumentUnderstanding.Activities package provides 3 new activities:
- Create Classification Validation Task
- Create Classification Validation Task and Wait
- Wait for Classification Validation Task and Resume
which have the purpose of creating Classification Tasks in Action Center, so business users can validate the result of a classification as well as manually split documents. The result of the split documents can be found as part of the DocumentData#SubDocuments property. Read more about them in their official documentation.
Support validator notes at field level
Introducing Validator Notes, a new feature that empowers users to add contextual notes during document validation in the Validation Station. With Validator Notes, users can articulate their decisions, specify additional checks, or flag specific conditions for downstream processing. Whether it’s acknowledging missing but acceptable elements like a signature field or indicating the need for secondary validation due to discrepancies in extracted data, the Validator Notes feature provides valuable insights for subsequent steps in the business process. The feature supports the configuration of following settings in the Taxonomy Manager:
- Read-only: view the validator notes set on the ExtractionResult field in Validation Station, as a message. If disabled (default state), the human validator can also edit that note in Validation Station and thus communicate back to the robot information about the decision taken.
- Text: validator note is displayed as a text message in Validation Station. The human validator can view, edit, or add a maximum of 200 characters message in Validation Station.
- Options: allows for the configuration of options that the human validator can view or choose from
The settings will be persisted on the Extraction Result and accessible in the Validation Station. Read more about configuring the feature as well as using it in the Validation Station in the corresponding documentation.
Auto Validation
The Auto Validation feature provides a streamlined approach to validating the extracted data based on different business rules. In its first version, the feature will compare extraction results provided by the used extraction methods (i.e. a specialized extraction models, form extractions, and regex extractions) falling below a certain confidence score, against the result of the generative extractor: if both results are equal, then the extraction confidence is boosted to the set confidence threshold. By boosting the confidences for matching values and lowering them for discrepancies, RPA developers can optimize workflows, reduce manual efforts, and minimize false positives.
The feature is available in preview in Cloud APIs, as part of the Data Extraction Scope in the IntelligentOCR package as well as part of the Extract Document Data Activity in the DocumentUnderstanding.Activities package.
Anil_G
ashokkarale
CHEN-CC
ahsan_khan
rkelchuri
Mayra_Alejandra_Paredes_C
lulachee