# US 2025/0148537 A1 · Automated Financial Reporting Error Detector using NLP and ML

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

- Number: US 2025/0148537 A1
- Document ID: US20250148537A1
- Status: Published application
- Kind: A1
- Application number: 18/502,009
- Filed: 2023-11-04
- Published: 2025-05-08
- Earliest claimed: 2023-11-04
- Inventors: Joshua Michael Maluchnik
- Web page: https://patentgrep.com/patent/US20250148537A1

## CPC

- G06Q 40/06: Asset management; Financial planning or analysis
- G06F 40/205: Parsing
- G06F 40/40: Processing or translation of natural language (natural language analysis G06F40/20; semantic analysis G06F40/30)
- 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
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

The present invention is an advanced software platform that masterfully integrates natural language processing (NLP) and machine learning (ML) to achieve superior accuracy in automatically identifying discrepancies in financial documents. Utilizing NLP, the software meticulously analyzes the language within financial texts while ML algorithms enhance its ability to detect anomalies and errors that might be missed during manual checks. This system caters to financial entities, auditors, and companies, significantly improving review processes and guarding against the risks of human oversight. Consequently, it assures the reliability of financial records, bolstering trust in organizations' financial reports, and ensuring adherence to regulatory norms. This breakthrough epitomizes the synergy of financial expertise and cutting-edge technology, redefining standards in financial data verification.

## Claims

### Claim 1

A software solution capable of automatically detecting anomalies in financial reports using natural language processing.

### Claim 2 (depends on 1)

A method where the parsed financial data from claim 1 is converted into a structured format for easier analysis.

### Claim 3

A system designed with a user interface that allows for easy uploading and reviewing of financial reports.

### Claim 4 (depends on 1)

The system as claimed in claim 1, wherein machine learning models are employed to enhance detection accuracy based on historical data.

### Claim 5 (depends on 1)

The system as claimed in claim 1, which offers users a summarized report of detected inconsistencies.

### Claim 6 (depends on 3)

The system as claimed in claim 3, that incorporates features for uploading financial reports via various methods including API, web services, and SFTP.

### Claim 7 (depends on 2)

The method as claimed in claim 2, wherein financial data is classified and segregated according to predefined categories.

### Claim 8 (depends on 1)

The system as claimed in claim 1, which includes an adaptive learning mechanism that refines anomaly detection processes over time.

### Claim 9 (depends on 1)

The system as claimed in claim 1, where the anomaly detection is facilitated by a combination of both rule-based and probabilistic approaches.

### Claim 10 (depends on 3)

The system as claimed in claim 3, including feedback mechanisms allowing users to rectify and address the detected anomalies.

### Claim 11 (depends on 1)

The system as claimed in claim 1, with a built-in alert system that notifies users upon detection of critical inconsistencies in financial data.

### Claim 12 (depends on 3)

The system as claimed in claim 3, that provides user access controls and role-based permissions for various functionalities.

### Claim 13 (depends on 2)

The system as claimed in claim 2, that employs normalization and standardization techniques for the input financial data.

### Claim 14 (depends on 1)

The system as claimed in claim 1, which integrates with external financial databases or third-party interfaces for additional data validation.

### Claim 15 (depends on 3)

The system as claimed in claim 3, with encryption and secure data transmission protocols ensuring data safety.

### Claim 16 (depends on 2)

The method as claimed in claim 2, where the structured format assists in comparative financial analysis over different periods.

### Claim 17 (depends on 4)

The system as claimed in claim 4, that maintains a repository of historical discrepancies to facilitate the machine learning model's training.

### Claim 18 (depends on 6)

The system as claimed in claim 6, that automates the ingestion of financial data at predefined intervals.

### Claim 19 (depends on 1)

The system as claimed in claim 1, which can be deployed across various platforms including cloud, on-premises, and hybrid environments.

### Claim 20 (depends on 2)

The method as claimed in claim 2, where additional metadata is generated to provide context to the structured financial data.
