# US 2026/0064668 A1 · System and method for augmenting large language models with graph knowledge generated by universal modeling of datasets

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

- Number: US 2026/0064668 A1
- Document ID: US20260064668A1
- Status: Published application
- Kind: A1
- Application number: 19/302,931
- Filed: 2025-08-18
- Published: 2026-03-05
- Earliest claimed: 2024-09-04
- Assignees: Oracle International Corporation
- Applicants: Oracle International Corporation
- Inventors: Kalyan Beemanapalli; Gustavo Alejandro Cruz Silva; Huanxin Ding; Luxi Wang; Zijiao Wang
- Web page: https://patentgrep.com/patent/US20260064668A1

## CPC

- G06F 16/2423: {Interactive query statement specification based on a database schema}
- G06F 16/243: {Natural language query formulation}
- G06F 16/288: {Entity relationship models}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Embodiments described herein are generally related to data analytics environments, and are particularly directed to systems and methods for augmenting large language models with graph knowledge generated by universal modelling of datasets. In accordance with an embodiment, a method for augmenting large language models with graph knowledge generated by universal modeling of datasets, is provided. The method can provide, by a computer including one or more processors, access to a data analytics environment. The method can create a graph schema associated with a dataset of the data analytics environment. The method can receive a query associated with the dataset of the data analytics environment. The method can receive, at a large language model, a parsed version of the query, together with the graph schema. The method can, based upon the received parsed query and the graph schema, generate, by the large language model, a graph query.

## Claims

### Claim 1

A system for augmenting large language models with graph knowledge generated by universal modelling of datasets, comprising:
a computer including one or more processors, that provides access to a data analytics environment; and
wherein a graph schema associated with a dataset of the data analytics environment is created;
wherein a query associated with the dataset of the data analytics environment is received;
wherein a parsed version of the query, together with the graph schema, are received at a large language model; and
wherein, based upon the received parsed query and the graph schema, the large language model generates a graph query.

### Claim 2 (depends on 1)

The system of claim 1, wherein the dataset comprises a schema comprising one of a relational schema and a start schema.

### Claim 3 (depends on 2)

The system of claim 2, wherein the generated graph query is run against a graph database.

### Claim 4 (depends on 3)

The system of claim 3, wherein results of the generated graph query being run against the graph database are received at another large language model; and
wherein a natural language version of the results of the generated graph query being run against the database is received from the large language model.

### Claim 5 (depends on 4)

The system of claim 4, wherein the received query comprises a natural language format; and
wherein the natural language version of the results of the generated graph query is caused to be displayed via one or more user interfaces.

### Claim 6 (depends on 5)

The system of claim 5, wherein the received query is directed to at least a portion of a displayed data visualization associated with the dataset.

### Claim 7 (depends on 1)

The system of claim 1, wherein the graph schema associated with the dataset comprises data from a plurality of other datasets.

### Claim 8

A method for augmenting large language models with graph knowledge generated by universal modelling of datasets, comprising:
providing, by a computer including one or more processors, access to a data analytics environment; and
creating a graph schema associated with a dataset of the data analytics environment;
receiving a query associated with the dataset of the data analytics environment;
receiving, at a large language model, a parsed version of the query, together with the graph schema; and
based upon the received parsed query and the graph schema, generating, by the large language model, a graph query.

### Claim 9 (depends on 8)

The method of claim 8, wherein the dataset comprises a schema comprising one of a relational schema and a start schema.

### Claim 10 (depends on 9)

The method of claim 9, wherein the generated graph query is run against a graph database.

### Claim 11 (depends on 10)

The method of claim 10, wherein results of the generated graph query being run against the graph database are received at another large language model; and
wherein a natural language version of the results of the generated graph query being run against the database is received from the large language model.

### Claim 12 (depends on 11)

The method of claim 11, wherein the received query comprises a natural language format; and
wherein the natural language version of the results of the generated graph query is caused to be displayed via one or more user interfaces.

### Claim 13 (depends on 12)

The method of claim 12, wherein the received query is directed to at least a portion of a displayed data visualization associated with the dataset.

### Claim 14 (depends on 8)

The method of claim 8, wherein the graph schema associated with the dataset comprises data from a plurality of other datasets.

### Claim 15

A non-transitory computer readable storage medium having instructions thereon for augmenting large language models with graph knowledge generated by universal modelling of datasets, which when read and executed cause a computer to perform steps comprising:
providing, by the computer, the computer including one or more processors, access to a data analytics environment; and
creating a graph schema associated with a dataset of the data analytics environment;
receiving a query associated with the dataset of the data analytics environment;
receiving, at a large language model, a parsed version of the query, together with the graph schema; and
based upon the received parsed query and the graph schema, generating, by the large language model, a graph query.

### Claim 16 (depends on 15)

The non-transitory computer readable storage medium of claim 15, wherein the dataset comprises a schema comprising one of a relational schema and a start schema.

### Claim 17 (depends on 16)

The non-transitory computer readable storage medium of claim 16, wherein the generated graph query is run against a graph database.

### Claim 18 (depends on 17)

The non-transitory computer readable storage medium of claim 17, wherein results of the generated graph query being run against the graph database are received at another large language model; and
wherein a natural language version of the results of the generated graph query being run against the database is received from the large language model.

### Claim 19 (depends on 18)

The non-transitory computer readable storage medium of claim 18, wherein the received query comprises a natural language format; and
wherein the natural language version of the results of the generated graph query is caused to be displayed via one or more user interfaces.

### Claim 20 (depends on 19)

The non-transitory computer readable storage medium of claim 19, wherein the received query is directed to at least a portion of a displayed data visualization associated with the dataset.
