User-defined automated document feature modeling, extraction and optimization

Patent metadata

Field	Value
Publication number	US-11048762-B2
Application number	US-201916299128-A
Country	US
Kind code	B2
Filing date	Mar 11, 2019
Priority date	Mar 16, 2018
Publication date	Jun 29, 2021
Grant date	Jun 29, 2021

Abstract

Provided herein are systems and methods for user-defined automated document feature modeling, extraction and optimization. In the present disclosure, an end user of an automated document review system can customize and create new data models applicable to a set of focus documents. In addition, an end user of the automated document review system can customize and create new extraction rules applicable to text extraction from the set of focus documents. The user-defined edits to the data model and extraction rules can be further tested in a staging environment, and tested against a ground truth set of documents, before being widely applied to other relevant documents.

First claim

What is claimed is:

1. A computer-implemented method for optimizing automated document feature extraction, the method comprising:
receiving a selection from a user of a data model to be updated or a new data model to be created;
providing the user with a graphical user interface to edit one or more fields within the selected data model;
receiving an edit from the user, via the graphical user interface, to the one or more fields within the selected data model;
generating an updated data model based on the received edit from the user;
assigning each of the data model and the updated data model a unique version;
updating a staging environment for an automated document feature extraction system;
performing automated document feature extraction utilizing the data model and the updated data model on a focus set of documents;
displaying output of the automated document feature extraction for each of the data model and the updated data model;
publishing the data model and the updated data model in a staging environment for testing and verification, to ensure that data model changes based on the received edit from the user are operating as intended;
displaying, in the staging environment, the output of the automated document feature extraction from the data model and the updated data model so that a user can verify that the automated document feature extraction of the updated data model does not disrupt previous document field modeling and extraction using the data model, in an undesirable manner; and
publishing the updated data model in the automated document feature extraction system to enhance automated document feature extraction within the focus set of documents and a second set of additional documents.

2. The method of claim 1 , further comprising:
verifying that the automated document feature extraction utilizing the updated data model passes a quality check on a ground truth set of documents.

3. The method of claim 1 , wherein the received edit from the user to the one or more fields within the selected data model is a creation of a new field within the selected data model.

4. The method of claim 1 , wherein the received edit from the user to the one or more fields within the selected data model is a modification or deletion of an existing field within the selected data model.

5. The method of claim 1 , wherein the received edit from the user to the one or more fields within the selected data model comprises reordering of at least one field within a group.

6. The method of claim 1 , wherein the received edit from the user to the one or more fields within the selected data model comprises adding or removing at least one field from a group.

7. A computer-implemented method for optimizing automated document feature extraction, the method comprising:
receiving a selection from a user of a data model to be updated or a new data model to be created;
providing the user with a graphical user interface to edit one or more extraction rules within the selected data model;
receiving an edit from the user, via the graphical user interface, to the one or more extraction rules within the selected data model;
generating an updated data model based on the received edit from the user;
assigning each of the data model and the updated data model a unique version;
updating a staging environment for an automated document feature extraction system;
performing automated document feature extraction utilizing the data model and the updated data model on a focus set of documents;
displaying output of the automated document feature extraction for each of the data model and the updated data model;
publishing the data model and the updated data model in a staging environment for testing and verification, to ensure that data model changes based on the received edit from the user are operating as intended;
displaying, in the staging environment, the output of the automated document feature extraction from the data model and the updated data model so that a user can verify that the automated document feature extraction of the updated data model does not disrupt previous document field modeling and extraction using the data model, in an undesirable manner;
publishing the updated data model in the automated document feature extraction system to enhance automated document feature extraction within the focus set of documents and a second set of additional documents.

8. The method of claim 7 , further comprising:
verifying that the automated document feature extraction utilizing the updated data model passes a quality check on a ground truth set of documents.

9. The method of claim 7 , wherein the received edit from the user to the one or more extraction rules is a reordering of at least two extraction rules.

10. The method of claim 7 , wherein the received edit from the user to the one or more extraction rules is creation of a new extraction rule.

11. The method of claim 7 , wherein the received edit from the user to the one or more extraction rules is modification or deletion of one or more existing extraction rules.

12. The method of claim 7 , further comprising:
displaying to the user, via the graphical user interface, an amount of substitutions, deletions, and insertions made by the automated document feature extraction system as a result of the updated data model.

13. A system, comprising:
a processor; and
a memory for storing executable instructions, the processor executing the instructions to:
receive a selection from a user of a data model to be updated or a new data model to be created;
provide the user with a graphical user interface to edit one or more fields within the selected data model;
receive an edit from the user, via the graphical user interface, to the one or more fields within the selected data model;
generate an updated data model based on the received edit from the user;
assign each of the data model and the updated data model a unique version;
update a staging environment for an automated document feature extraction system;
perform automated document feature extraction utilizing the data model and the updated data model on a focus set of documents;
publish the data model and the updated data model in a staging environment for testing and verification, to ensure that data model changes based on the received edit from the user are operating as intended;
provide, in the staging environment, the output of the automated document feature extraction from the data model and the updated data model so that a user can verify that the automated document feature extraction of the updated data model does not disrupt previous document field modeling and extraction using the data model, in an undesirable manner;
publish the updated data model in the automated document feature extraction system to enhance automated document feature extraction within the focus set of documents and a second set of additional documents.

14. The system of claim 13 , wherein the processor further executes the instructions to:
verify that the automated document feature extraction utilizing the updated data model passes a quality check on a ground truth set of documents.

15. The system of claim 13 , wherein the received edit from the user to the one or more fields within the selected data model is a creation of a new field within the selected data model.

16. The system of claim 13 , wherein the received edit from the user to the one or more fields within the selected data model is a modification or deletion of an existing field within the selected data model.

17. The system of c

Assignees

• Open Text Holdings Inc

Inventors

• Puzicha Jan
• Stadermann Jan
• Muppala Chaitanya
• Yanamandra Sangeetha
• Deshpande Ketan

Classifications

• G06V10/993 Primary

  Evaluation of the quality of the acquired pattern · CPC title

• G06V30/40

  Document-oriented image-based pattern recognition · CPC title

• G06F16/93 Primary

  Document management systems · CPC title

• G06F40/166

  Editing, e.g. inserting or deleting · CPC title

• G06K9/2054

  Physics · mapped topic