IP Library Granted Patent US 11,769,008
Granted Patent B2
US 11,769,008 · App. 17/817,997 · Granted Sep 26, 2023

Predictive analysis systems and methods using machine learning

Inventors: Wacey T. Richards (Midway, UT); Michael E. Kiemel (Lehi, UT); Stephen W. Shillingford (Salt Lake City, UT); Samuel Z. Shillingford (Salt Lake City, UT); Damon A. Darais (Pleasant Grove, UT); Joseph M. Wood (Broadlands, VA); Robert D. Bailey (Saratoga Springs, UT); Matthew Valley (American Fork, UT); Stewart A. Sintay (Lehi, UT)
G06F40/205G06F18/2178G06N20/00G06V30/414G06V30/416
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,769,008
App. No.
17/817,997
Granted
Sep 26, 2023
Kind
B2
Abstract

Systems and methods may utilize a predictive analysis model to analyze a contract or other document. A system may parse a document and/or a repository of information associated with the document. The system may identify one or more terms in the document and corresponding terms in the repository. The system may determine a difference parameter between a first term extracted from the document and a second term extracted from the repository. The system may determine whether the difference between the first term and the second term, represented by the difference parameter, is likely to be acceptable to the user using a predictive analysis model. The system may report a validation parameter indicating a level of acceptability associated with the difference. User feedback on the accuracy of the predictive analysis model is used to train, modify, and improve the predictive analysis model.

Claims (67)

1. A system, comprising:

a term extraction module to:

extract a first term from a document, and

extract a second term from a repository of terms based at least in part on the first term extracted from the document;

a comparison module to:

identify a difference parameter between the first term extracted from the document and the second term extracted from the repository of terms, and

determine whether the difference parameter between the first term and the second term satisfies a threshold based on a predictive analysis model; and

a validation module to identify a validation parameter based on the determination of whether the difference parameter satisfies the threshold based on the predictive analysis model, wherein the validation parameter indicates a quantified level of acceptability of a difference between the first term and the second term.

2. The system of claim 1 , wherein the document comprises one or more of a deal document, a contract document, a transaction document, a negotiation document, an agreement document, and a legal document.

3. The system of claim 1 , further comprising:

a feedback module to receive feedback from a user to validate the validation parameter generated using the predictive analysis model;

a predictive analysis module to dynamically modify the predictive analysis model when the received feedback indicates that the comparison module assigned an incorrect difference parameter; and

a machine learning manager to perform machine learning using the difference parameter, the validation parameter, and the feedback from the user to adjust the predictive analysis model.

4. The system of claim 3 , further comprising a confidence score manager to generate a confidence score associated with the validation parameter using the predictive analysis model that indicates a likelihood that the validation parameter is correct for the difference parameter between the first term and the second term, and

wherein the feedback module requests the feedback from the user at least partially in response to the generated confidence score.

5. The system of claim 1 , further comprising:

a feedback module to receive feedback from a user to validate the validation parameter generated using the predictive analysis model; and

a predictive analysis module to:

dynamically modify the predictive analysis model when the received feedback indicates that the comparison module assigned an incorrect difference parameter, and

implement a validated difference parameter comparison in which the difference parameter is compared to one or more difference parameters of validated terms stored in a database.

6. The system of claim 1 , further comprising:

a feedback module to receive feedback from a user to validate the validation parameter generated using the predictive analysis model; and

a predictive analysis module to dynamically modify the predictive analysis model when the received feedback indicates that the comparison module assigned an incorrect difference parameter,

wherein the predictive analysis module is configured to determine a risk profile of the difference parameter based on a risk evaluation of the difference parameter.

7. The system of claim 1 , wherein the threshold is an acceptance threshold, a validation threshold, a rejection threshold, a no-prediction threshold, or a combination thereof.

8. A system, comprising:

a term extraction module to:

extract a first term from a first document, and

extract a second term from a second document based at least in part on the first term extracted from the first document;

a comparison module to:

identify a difference parameter between the first term extracted from the first document and the second term extracted from the second document, and

determine whether the difference parameter between the first term and the second term satisfies a threshold based on a predictive analysis model; and

a validation module to identify a validation parameter based on the determination of whether the difference parameter satisfies the threshold based on the predictive analysis model, wherein the validation parameter indicates a quantified level of acceptability of a difference between the first term and the second term.

9. The system of claim 8 , wherein the first document comprises one or more of a deal document, a contract document, a transaction document, a negotiation document, an agreement document, and a legal document.

10. The system of claim 8 , further comprising a data repository to hold a plurality of documents, and

wherein the term extraction module retrieves the second term from the second document from within the data repository.

11. A non-transitory computer-readable medium storing code, the code comprising instructions executable by a processor to:

extract a first term from a document;

extract a second term from a repository of terms based at least in part on the first term extracted from the document;

determine a difference parameter based on a difference between the first term extracted from the document and the second term extracted from the repository of terms;

determine whether the difference parameter satisfies a threshold using a predictive analysis model; and

generate a validation parameter that indicates a level of acceptability of a difference between the first term and the second term.

12. The non-transitory computer-readable medium of claim 11 , wherein the document comprises one or more of a deal document, a contract document, a transaction document, a negotiation document, an agreement document, and a legal document.

13. A non-transitory computer-readable medium storing code, the code comprising instructions executable by a processor to:

extract a first term from a first document;

extract a second term from a second document based at least in part on the first term extracted from the first document;

determine a difference parameter based on a difference between the first term extracted from the first document and the second term extracted from the second document;

determine whether the difference parameter satisfies a threshold using a predictive analysis model; and

generate a validation parameter that indicates a level of acceptability of a difference between the first term and the second term.

14. The non-transitory computer-readable medium of claim 13 , wherein the first document comprises one or more of a deal document, a contract document, a transaction document, a negotiation document, an agreement document, and a legal document.

15. The non-transitory computer-readable medium of claim 13 , wherein the instructions executable by the processor to extract the second term from the second document comprise extracting the second document from a data repository of a plurality of documents.

16. The non-transitory computer-readable medium of claim 13 , wherein the instructions are further executable to:

store, in a database:

the difference parameter between the first term and the second term, and

the validation parameter; and

perform machine learning on information stored in the database to adjust results of the predictive analysis model, wherein generating the validation parameter using the predictive analysis model is based at least in part on performing the machine learning.

17. The non-transitory computer-readable medium of claim 13 , wherein the instructions are further executable to:

categorize the first term extracted from the first document; and

categorize the second term extracted from the second document, wherein identifying the difference parameter is based at least in part on the categorizations of the first term and the second term.

18. The non-transitory computer-readable medium of claim 17 , wherein the instructions are further executable to:

determine a validation criterion from a set of validation criteria to apply to the difference parameter based at least in part on categorizing the first term and the second term; and

apply the validation criterion to the difference parameter, wherein determining whether the difference parameter satisfies the threshold is based at least in part on applying the validation criterion.

19. The non-transitory computer-readable medium of claim 13 , wherein the instructions are further executable to:

identify that a second difference parameter between validated terms is similar to the difference parameter between the first term and the second term, wherein the second difference parameter is selected from a database of past validated terms; and

identify a second validation parameter associated with the second difference parameter, wherein determining the validation parameter using the predictive analysis model is based at least in part on identifying the second validation parameter associated with the second difference parameter.

20. The non-transitory computer-readable medium of claim 13 , wherein the instructions are further executable to:

correlate the first term extracted from the first document with the second term extracted from the second document, wherein identifying the difference parameter is based at least in part on correlating the first term and the second term.

Assignments (3)
RELEASE OF SECURITY INTEREST Recorded Nov 3, 2025
From: FIRST-CITIZENS BANK & TRUST COMPANY
To: DEEPSEE.AI INC.
Reel/Frame 072766/0875 →
SECURITY INTEREST Recorded Dec 28, 2023
From: DEEPSEE.AI INC.
To: FIRST-CITIZENS BANK & TRUST COMPANY
Reel/Frame 065967/0789 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 16, 2022
From: WOOD, JOSEPH M.; BAILEY, ROBERT D.; VALLEY, MATTHEW; SINTAY, STEWART A.; SHILLINGFORD, STEPHEN W.; RICHARDS, WACEY T.; DARAIS, DAMON A.; KIEMEL, MICHAEL E.; SHILLINGFORD, SAMUEL Z.
To: DEEPSEE.AI INC.
Reel/Frame 060821/0573 →
Continuity (4)
Continuation 17099566 · Nov 16, 2020
Continuation 16510644 · Jul 12, 2019
Provisional Application 62698106 · Jul 14, 2018
Related Publication 20230196012A1 · Jun 22, 2023