Dynamic validation of system transactions based on machine learning analysis

Patent metadata

Field	Value
Publication number	US-10706418-B2
Application number	US-201815917463-A
Country	US
Kind code	B2
Filing date	Mar 9, 2018
Priority date	Mar 9, 2018
Publication date	Jul 7, 2020
Grant date	Jul 7, 2020

Abstract

Systems and methods for applying machine learning to dynamically validate a sales transaction document created by a user in a computing system are provided. Data comprising the sales transaction document is received. A machine learning model is applied to the sales transaction document to verify that the sales transaction document meets at least one compliance standard. The user is alerted if the sales transaction document does not meet the at least one compliance standard. The maching learning model is generated by: receiving first sales transaction data from a database; determining patterns based on the first sales transaction data, wherein the patterns indicate that corrective data was created to compensate for at least one error in original data, the at least one error indicating that the original data did not meet at least one compliance standard; and generating the machine learning model based on the determined patterns.

First claim

What is claimed is:

1. A computer-implemented method for applying machine learning to dynamically validate a sales transaction document created by a user in a computing system, the method comprising:
receiving data comprising the sales transaction document;
applying a machine learning model to the sales transaction document to verify that the sales transaction document meets at least one compliance standard;
dynamically generating, using the machine learning model, a probability of compliance with the at least one compliance standard based on first sales transaction data;
alerting the user if the sales transaction document does not meet the at least one compliance standard;
receiving second sales transaction data from the database; and
adjusting the machine learning model based on the second sales transaction data,
wherein the machine learning model is trained by:
receiving the first sales transaction data from a database;
training the machine learning model using the first sales transaction data;
determining and subsequently analyzing patterns based on the first sales transaction data, wherein the patterns indicate that corrective data was created to compensate for at least one error in original data, the at least one error indicating that the original data did not meet at least one compliance standard, wherein the corrective data comprises a change to a noncompliant sales transaction document in the first sales transaction data; and
training the machine learning model, after the training using the first sales transaction data, based on the determined patterns.

2. The computer-implemented method of claim 1 , wherein the compliance standard is a tax law issued by a government entity.

3. The computer-implemented method of claim 2 , wherein the at least one error indicates that an incorrect amount was paid for a tax.

4. The computer-implemented method of claim 2 , wherein the at least one error indicates that a tax was not paid.

5. The computer-implemented method of claim 1 , wherein the corrective data comprises a new sales transaction document that reverses a noncompliant sales transaction document in the first sales transaction data.

6. The computer-implemented method of claim 1 , wherein the sales transaction document includes at least one of a purchase order and a sales order.

7. The computer-implemented method of claim 1 , wherein alerting the user further comprises notifying the user of a probability that the sales transaction document does not meet the at least one compliance standard.

8. The computer implemented method of claim 1 , further comprising statically validating whether the sales transaction document complies with at least one of a software requirement, an internal company standard, or an accounting standard.

9. The computer implemented method of claim 1 , wherein the corrective data comprises feedback data having a plurality of noncompliant sales transactions.

10. The computer implemented method of claim 1 , wherein the receiving, the applying, the dynamically generating, the alerting, and the adjusting are executed in an in-memory database.

11. A system for applying machine learning to dynamically validate a sales transaction document created by a user in a computing system, the system comprising:
one or more data processors having memory storing instructions, which when executed result in operations comprising:
receiving data comprising the sales transaction document;
applying a machine learning model to the sales transaction document to verify that the sales transaction document meets at least one compliance standard;
dynamically generating, using the machine learning model, a probability of compliance with the at least one compliance standard based on first sales transaction data;
alerting the user if the sales transaction document does not meet the at least one compliance standard;
receiving second sales transaction data from the database; and
adjusting the machine learning model based on the second sales transaction data,
wherein the machine learning model is trained by:
receiving the first sales transaction data from a database;
training the machine learning model using the first sales transaction data;
determining and subsequently analyzing patterns based on the first sales transaction data, wherein the patterns indicate that corrective data was created to compensate for at least one error in original data, the at least one error indicating that the original data did not meet at least one compliance standard, wherein the corrective data comprises a change to a noncompliant sales transaction document in the first sales transaction data; and
training the machine learning model, after the training using the first sales transaction data, based on the determined patterns.

12. The system of claim 11 , wherein the compliance standard is a tax law issued by a government entity.

13. The system of claim 12 , wherein the at least one error indicates that an incorrect amount was paid for a tax.

14. The system of claim 12 , wherein the at least one error indicates that a tax was not paid.

15. The system of claim 11 , wherein the corrective data comprises a new sales transaction document that reverses a noncompliant sales transaction document in the first sales transaction data.

16. The system of claim 11 , wherein the sales transaction document includes at least one of a purchase order and a sales order.

17. The system of claim 11 , wherein alerting the user further comprises notifying the user of a probability that the sales transaction document does not meet the at least one compliance standard.

18. The system of claim 11 , further comprising an in-memory database.

19. A non-transitory computer readable storage medium storing one or more programs configured to be executed by one or more data processors, the one or more programs comprising instructions for applying machine learning to dynamically validate a sales transaction document created by a user in a computing system, the instructions comprising:
receiving data comprising the sales transaction document;
applying a machine learning model to the sales transaction document to verify that the sales transaction document meets at least one compliance standard;
dynamically generating, using the machine learning model, a probability of compliance with the at least one compliance standard based on first sales transaction data;
alerting the user if the sales transaction document does not meet the at least one compliance standard;
receiving second sales transaction data from the database; and
adjusting the machine learning model based on the second sales transaction data,
wherein the machine learning model is trained by:
receiving the first sales transaction data from a database;
training the machine learning model using the first sales transaction data;
determining and subsequently analyzing patterns based on the first sales transaction data, wherein the patterns indicate that corrective data was created to compensate for at least one error in original data, the at least one error indicating that the original data did not meet at least one compliance standard, wherein the corrective data comprises a change to a noncompliant sales transaction document in the first sales transaction data; and
training the machine learning model, after the training using the first sales transaction data, based on the determined patterns.

20. The non-transitory computer readable storage medium of claim 19 , wherein the receiving, th

Assignees

• Sap Se

Inventors

• Muller Andre

Classifications

• G06Q40/10

  Tax strategies · CPC title

• G06Q20/40 Primary

  Authorisation, e.g. identification of payer or payee, verification of customer or shop credentials; Review and approval of payers, e.g. check credit lines or negative lists · CPC title

• G06N20/00

  Machine learning · CPC title

• G07D7/00

  Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency · CPC title

• G06Q20/042

  characterized in that the payment protocol involves at least one cheque · CPC title