# US 12,118,002 B2 · Data processing system

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,118,002 B2
- Document ID: US12118002B2
- Status: Granted
- Kind: B2
- Application number: 17/782,634
- Filed: 2020-12-04
- Published: 2023-01-12
- Granted: 2024-10-15
- Earliest claimed: 2019-12-06
- Applicants: Ernst & Young GmbH
- Inventors: Till Rainer Heckschen; Michael Wiese
- Web page: https://patentgrep.com/patent/US12118002B2

## CPC

- G06F 16/24553: {of query operations}
- G06Q 40/00: Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR

## Abstract

Disclosed is a data processing system comprising a database arrangement and at least one processor coupled in communication with the database arrangement. The at least one processor is configured to obtain, from the database arrangement, labelled training data; process 5 the labelled training data using at least one machine learning algorithm to generate a data processing model; obtain, from the database arrangement or a user device, at least one test document; process the at least one test document using the data processing model to extract key information from the at least one test document; and compare the 10 extracted key information with reference key information to identify whether or not there exist inconsistencies between the extracted key information and the reference key information.

## Claims

### Claim 1

A data processing system comprising:
a database arrangement; and
at least one processor coupled in communication with the database arrangement, wherein the at least one processor is configured to:
obtain, from the database arrangement, labelled training data;
process the labelled training data using at least one machine learning algorithm to generate a data processing model;
obtain, from the database arrangement or a user device, at least one test document;
process the at least one test document using the data processing model to extract key information from the at least one test document; and
compare the extracted key information with reference key information to identify if inconsistencies exist between the extracted key information and the reference key information,
wherein the at least one processor is further configured to represent, via an interactive user interface, a summary of the comparison between the extracted key information and the reference key information, and wherein the summary is indicative of a percentage degree of similarity between the extracted key information and the reference key information.

### Claim 2 (depends on 1)

The data processing system of claim 1, wherein the labelled training data comprises a plurality of training documents and a training target document, the training target document comprising labelled key information extracted from the plurality of training documents.

### Claim 3 (depends on 2)

The data processing system of claim 2, wherein the at least one machine learning algorithm comprises at least one of: a decision tree-based algorithm, a gradient boosting algorithm, a neural network-based supervised learning algorithm.

### Claim 4 (depends on 3)

The data processing system of claim 3, wherein the at least one test document comprises a plurality of test documents, and the at least one processor is configured to employ parallel processing for processing the plurality of test documents using the data processing model.

### Claim 5 (depends on 1)

The data processing system of claim 1, wherein the at least one test document comprises a plurality of test documents, and the at least one processor is configured to employ parallel processing for processing the plurality of test documents using the data processing model.

### Claim 6 (depends on 5)

The data processing system of claim 5, wherein the labelled training data comprises labelled financial data, the at least one test document comprises at least one financial document, the extracted key information comprises financial data extracted from the at least one financial document, and the reference key information comprises reference financial data posted by an entity to be audited in a general ledger or a sub-ledger of the entity.

### Claim 7 (depends on 1)

The data processing system of claim 1, wherein the at least one machine learning algorithm comprises at least one of: a decision tree-based algorithm, a gradient boosting algorithm, a neural network-based supervised learning algorithm.

### Claim 8 (depends on 1)

The data processing system of claim 1, wherein the labelled training data comprises labelled financial data, the at least one test document comprises at least one financial document, the extracted key information comprises financial data extracted from the at least one financial document, and the reference key information comprises reference financial data posted by an entity to be audited in a general ledger or a sub-ledger of the entity.