# US 2024/0203149 A1 · System and method for domain aware document classification and information extraction from consumer documents

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

- Number: US 2024/0203149 A1
- Document ID: US20240203149A1
- Status: Published application
- Kind: A1
- Application number: 18/590,472
- Filed: 2024-02-28
- Published: 2024-06-20
- Earliest claimed: 2020-08-26
- Assignees: Informed S.A.
- Applicants: Informed, Inc.
- Inventors: Jatin Agrawal; Ashwin Kannan; Harshil Prajapati; Bharath Rengarajan; Eric Harvey; Emma Wei
- Web page: https://patentgrep.com/patent/US20240203149A1

## CPC

- G06V 30/416: Extracting the logical structure, e.g. chapters, sections or page numbers; Identifying elements of the document, e.g. authors
- G06F 40/226: Validation
- G06F 40/289: Phrasal analysis, e.g. finite state techniques or chunking
- G06F 40/30: Semantic analysis
- G06V 30/412: Layout analysis of documents structured with printed lines or input boxes, e.g. business forms or tables
- G06V 30/414: Extracting the geometrical structure, e.g. layout tree; Block segmentation, e.g. bounding boxes for graphics or text
- G06V 30/10: Character recognition
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A system and method for domain aware document classification and information extraction from consumer documents are disclosed. A particular embodiment is configured to: establish, by use of a data processor and a data network, a data connection with at least one applicant platform; receive an upload of documents from the applicant platform via the data network; classify each document as being of a particular document type; determine an information extraction strategy based on a document type classification of a particular document; and extract information from the particular document based on the information extraction strategy.

## Claims

### Claim 1

A computer-implemented method comprising:
establishing, by use of a data processor and a data network, a data connection with at least one applicant platform;
receiving an upload of documents from the applicant platform via the data network;
classifying each document as being of a particular document type;
determining an information extraction strategy based on a document type classification of a particular document, the information extraction strategy including performing extractions from the document based on the document type classification and a document context, the document context including an inference of document content drawn from document content extracted from a plurality of related documents, the document context further including applying domain-specific rules of a domain-aware business layer to bias document content extraction toward compliance with the domain-specific rules; and
extracting information from the particular document based on the information extraction strategy and the document context.

### Claim 2 (depends on 1)

The method of claim 1 including merging pages of the documents based on the document type classification.

### Claim 3 (depends on 1)

The method of claim 1 including performing a spatial phrase construction operation on the document.

### Claim 4 (depends on 1)

The method of claim 1 including performing feature extractions on the document, the extracted features including a plurality of sections of the document.

### Claim 5 (depends on 1)

The method of claim 1 including performing feature extractions on the document, the extracted features including a checkbox, a table, or a signature block.

### Claim 6 (depends on 1)

The method of claim 1 wherein the document context further includes use of applicant-stated information to make predictions regarding the document type classification or an entity's value.

### Claim 7 (depends on 1)

The method of claim 1 including using a trained machine learning model to extract information from the document.

### Claim 8 (depends on 1)

The method of claim 1 wherein the document context further includes use of approval terms of a loan application to make predictions regarding the document type classification.

### Claim 9 (depends on 1)

The method of claim 1 including performing major rotations or minor rotations of a document.

### Claim 10 (depends on 1)

The method of claim 1 including modifying a document type classification of a first document based on information extracted from a different second document.

### Claim 11

A system comprising:
a data processor;
a network interface, in data communication with the data processor, for communication on a data network; and
an information extraction system, executable by the data processor, to:
establish, by use of a data processor and a data network, a data connection with at least one applicant platform;
receive an upload of documents from the applicant platform via the data network;
classify each document as being of a particular document type;
determine an information extraction strategy based on a document type classification of a particular document, the information extraction strategy including performing extractions from the document based on the document type classification and a document context, the document context including an inference of document content drawn from document content extracted from a plurality of related documents, the document context further including applying domain-specific rules of a domain-aware business layer to bias document content extraction toward compliance with the domain-specific rules; and
extract information from the particular document based on the information extraction strategy and the document context.

### Claim 12 (depends on 11)

The system of claim 11 being further configured to merge pages of the documents based on the document type classification.

### Claim 13 (depends on 11)

The system of claim 11 being further configured to perform a spatial phrase construction operation on the document.

### Claim 14 (depends on 11)

The system of claim 11 being further configured to perform feature extractions on the document, the extracted features including a plurality of sections of the document.

### Claim 15 (depends on 11)

The system of claim 11 being further configured to perform feature extractions on the document, the extracted features including a checkbox, a table, or a signature block.

### Claim 16 (depends on 11)

The system of claim 11 being further configured to use a trained machine learning model to extract information from the document.

### Claim 17 (depends on 11)

The system of claim 11 wherein the document context further includes use of applicant-stated information to make predictions regarding the document type classification or an entity's value.

### Claim 18 (depends on 11)

The system of claim 11 wherein the document context further includes use of approval terms of a loan application to make predictions regarding the document type classification.

### Claim 19 (depends on 11)

The system of claim 11 being further configured to perform major rotations or minor rotations of a document.

### Claim 20 (depends on 11)

The system of claim 11 being further configured to modify a document type classification of a first document based on information extracted from a different second document.
