# US 2026/0064989 A1 · System and methods for classification of unstructured data using similarity metrics

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

- Number: US 2026/0064989 A1
- Document ID: US20260064989A1
- Status: Published application
- Kind: A1
- Application number: 19/380,589
- Filed: 2025-11-05
- Published: 2026-03-05
- Earliest claimed: 2022-09-12
- Assignees: Wells Fargo Bank, N.A.
- Applicants: Wells Fargo Bank, N.A.
- Inventors: Debashis Ghosh; Maria Chiarenza; Nathaniel D. Mollison; Matthew C. Howell
- Web page: https://patentgrep.com/patent/US20260064989A1

## CPC

- G06F 40/40: Processing or translation of natural language (natural language analysis G06F40/20; semantic analysis G06F40/30)
- G06F 40/174: Form filling; Merging
- G06F 40/284: Lexical analysis, e.g. tokenisation or collocates
- G06Q 40/03: Credit; Loans; Processing thereof
- 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

Systems, apparatuses, methods, and computer program products are disclosed for obtaining relevant data from an unstructured data source. An example method includes extracting relevant data that is intermixed with extraneous data using natural language processing. In order to do so, text from the unstructured data source may be tokenized and each token may be compared to an identifier associated with the relevant data. A similarity metric may be determined between each token and the identifier in order to classify tokens as similar or dissimilar to the identifier. All tokens classified as similar to the identifier may be aggregated in order to obtain relevant data.

## Claims

### Claim 1

A method for decision making that relies upon unstructured data, the method comprising:
receiving, from a client device, a request for services;
generating, based on the request for services, an identifier representing a first client;
identifying, by a data management circuitry of an information manager and based on the request for services, a data requirement event;
obtaining, by communication hardware of the information manager, the unstructured data in response to the data requirement event, the unstructured data comprising relevant data intermixed with extraneous data;
generating, by the data management circuitry, a first set of tokens representing the unstructured data;
determining, by the data management circuitry, a similarity metric for each token of the first set of tokens, each similarity metric indicating a likelihood that a corresponding token relates to the identifier;
identifying, by comparing the similarity metric for each token to a threshold value, a set of relevant tokens;
extracting, by the data management circuitry, the relevant data from the set of relevant tokens;
resolving, by the data management circuitry, the data requirement event using the relevant data; and
generating, by the data management circuitry and in response to resolving the data requirement event, a response to the request for services.

### Claim 2 (depends on 1)

The method of claim 1, wherein the identifier comprises a name of the first client.

### Claim 3 (depends on 2)

The method of claim 2, wherein the relevant data comprises indicators of financial income of the first client.

### Claim 4 (depends on 3)

The method of claim 3, wherein the extraneous data comprises indicators of financial income for a second client.

### Claim 5 (depends on 4)

The method of claim 4, wherein the unstructured data comprises a log of deposits to a financial account associated with the first client.

### Claim 6 (depends on 1)

The method of claim 1, wherein determining the similarity metric for each token of the first set of tokens comprises:
obtaining, by the data management circuitry, at least a first token of the first set of tokens;
obtaining, by the data management circuitry, the identifier representing the first client; and
determining, by the data management circuitry, at least a first similarity metric between the first token and the identifier representing the first client.

### Claim 7 (depends on 6)

The method of claim 6, wherein the first similarity metric is a Levenshtein's distance between the identifier representing the first client and the first token.

### Claim 8 (depends on 1)

The method of claim 1, wherein the request for services comprise a loan request.

### Claim 9 (depends on 8)

The method of claim 8, wherein generating the response to the request for services comprises:
generating, using the relevant data, a total income of the first client; and
determining, by comparing the total income of the first client to an institutional requirement for a loan, approval for the first client to receive the loan.

### Claim 10 (depends on 1)

The method of claim 1, further comprising:
receiving, from the client device, an incomplete service request form;
generating, using the relevant data, client data;
generating, by populating the client data into a service request form, a completed service request form; and
storing the completed service request form for approval.

### Claim 11 (depends on 10)

The method of claim 10, further comprising:
generating, based on the completed service request form, a graphical interface representing where the client data was populated in the completed service request form; and
sending the completed service request form and the graphical interface to the client device.

### Claim 12 (depends on 10)

The method of claim 10, further comprising:
generating, by an inference model processing the completed service request form, a response to the completed service request form; and
sending, to the client device, the response to the completed service request form.

### Claim 13 (depends on 1)

The method of claim 1, further comprising:
generating, using a data quality score manager and based on the unstructured data, a data quality score.

### Claim 14

An information manager for decision making that relies upon unstructured data, the information manager comprising:
communication hardware configured to receive from a client device a request for services;
a data management circuitry configured to:
identify a data requirement event based on the request for services, and
generate, based on the request for services, an identifier representing a first client,
wherein the communication hardware is further configured to obtain the unstructured data in response to the data requirement event, the unstructured data comprising relevant data intermixed with extraneous data,
wherein the data management circuitry is further configured to:
generate a first set of tokens representing the unstructured data,
determine a similarity metric for each token of the first set of tokens, each similarity metric indicating a likelihood that a corresponding token relates to the identifier,
identify, by comparing the similarity metric for each token to a threshold value, a set of relevant tokens,
extract the relevant data from the set of relevant tokens,
resolve the data requirement event using the relevant data, and
generate, in response to resolving the data requirement event, a response to the request for services.

### Claim 15 (depends on 14)

The information manager of claim 14, wherein the identifier comprises a name of the first client.

### Claim 16 (depends on 14)

The information manager of claim 14, wherein the relevant data comprises indicators of financial income of the first client.

### Claim 17 (depends on 14)

The information manager of claim 14, wherein the extraneous data comprises indicators of financial income for a second client.

### Claim 18 (depends on 14)

The information manager of claim 14, wherein the request for services comprise a loan request.

### Claim 19 (depends on 18)

The information manager of claim 18, wherein the data management circuitry is further configured to:
generate, using the relevant data, a total income of the first client; and
determine, by comparing the total income of the first client to an institutional requirement for a loan, approval for the first client to receive the loan.

### Claim 20

A computer program product for decision making that relies upon unstructured data, wherein the unstructured data comprises relevant data intermixed with extraneous data, the computer program product comprising at least one non-transitory computer-readable storage medium storing software instructions that, when executed, cause an apparatus to:
receive a request for services;
generate, based on the request for services, an identifier representing a first client;
identify, by a data management circuitry of an information manager and based on the request for services, a data requirement event;
obtain, by communication hardware of the information manager, the unstructured data in response to the data requirement event, the unstructured data comprising relevant data intermixed with extraneous data;
generate, by the data management circuitry, a first set of tokens representing the unstructured data;
determine, by the data management circuitry, a similarity metric for each token of the first set of tokens, each similarity metric indicating a likelihood that a corresponding toke relates to the identifier;
identify, by comparing the similarity metric for each token to a threshold value, a set of relevant tokens;
extract, by the data management circuitry, the relevant data from the set of relevant tokens;
resolve, by the data management circuitry, the data requirement event using the relevant data; and
generate, in response to resolving the data requirement event, a response to the request for services.
