Uipath Inc.

Paul Parau

Ioana Gligan

Remus Rusanu

Tony Y Tzeng

Tudor Serban

Tarun Singh

Tudor-Alexandru Carean

The present system and method relate generally to the field of Robotic Process Automation, particularly to a form data extractor for document processing. The system and method relate to a form extractor for document processing using RPA workflows that can be easily configured for different document types. The form extractor includes a set of templates for identifying the document type (classification) and extracting data from the documents. The templates can be configured, i.e., by the user, by defining the fields to be extracted and the position of the field on the document. The form extractor is resilient to changes in the position of the template on a page, as well as to scan rotation, size, quality, skew angle variations and file formats, thus allowing RPA processes to extract data from documents that need ingestion, independent of how they are created.

Form extractor

Systems and methods for classifying and splitting an electronic file into a plurality of extracted documents are provided. The electronic file is received. An initial portion of the electronic file is classified using a trained classifier and extracted from the electronic file as an extracted document associated with the classification. It is iteratively determined whether each respective next portion of the electronic file should be added to the extracted document until it is determined that the respective next portion should not be added to the extracted document. In response to determining that the respective next portion should be added to the extracted document, the respective next portion is extracted from the electronic file and added to the extracted document. In response to determining that the respective next portion should not be added to the extracted document, the classifying and the iteratively determining are repeated using the respective next portion as the initial portion. The extracted documents are output. The trained classifier can be trained to learn sets of word vectors and other relevant information associated with document classifications, in order to improve accuracy.

Self-improving document classification and splitting for document processing in robotic process automation

Systems and methods for executing a robotic process automation (RPA) workflow for document processing are provided. An input document is processed by a first robot executing one or more document processing activities of the RPA workflow. The document processing activities may include optical character recognition, digitization, classification, or data extraction. Execution of the RPA workflow is suspended by the first robot in response to a user validation activity of the RPA workflow. The user validation activity provides for user validation of the results of the one or more document processing activities. A user request that requests validation of the results from an end user is generated and the user request is transmitted to the end user. The execution of the RPA workflow is resumed by a second robot based on the validation received from the end user.

Growing community of inventors

Sibiu, Romania

Paul Parau