# US 2026/0064809 A1 · Generating cohesive explanations that communicate insights and patterns on multi-dimensional financial planning data

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

- Number: US 2026/0064809 A1
- Document ID: US20260064809A1
- Status: Published application
- Kind: A1
- Application number: 19/169,193
- Filed: 2025-04-03
- Published: 2026-03-05
- Earliest claimed: 2024-09-04
- Assignees: Oracle International Corporation
- Applicants: Oracle International Corporation
- Inventors: Manish Daga; Muthu Ranganathan; Selvarajan Kandasamy; Shivaranjan Gurunanjappa; Nitin Rawat
- Web page: https://patentgrep.com/patent/US20260064809A1

## CPC

- G06F 18/2178: {based on feedback of a supervisor}
- G06F 16/33295: {in dialogue systems}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Systems, articles, and computer-implemented methods are disclosed for generating natural language summaries of a multi-dimensional analysis of a detected anomaly within a member of multi-dimensional data by prompting a LLM with a prompt generated to include data about the anomaly in a manner understandable by the LLM. The prompt to the LLM includes a path to a member of the hierarchy containing an anomaly with a delimiter between the member and ancestor nodes. The delimiter allows the ancestral context of the member of the hierarchy to be understood by the LLM. The prompt also includes a metric defining a magnitude of the anomaly in relation to another value, such as an average, a value of the anomaly, a time corresponding to the anomaly, and one or more examples of other anomalies with included data about those anomalies matching the type of data provided for the detected anomaly.

## Claims

### Claim 1

A computer-implemented method comprising:
determining that analysis is enabled for a hierarchy of data of a multi-dimensional set of data, wherein analysis is disabled for one or more other hierarchies of data of the multi-dimensional set of data;
based at least in part on determining that the analysis is enabled, analyzing members of the hierarchy to detect a pattern or anomaly within the hierarchy;
in response to detecting the pattern or anomaly in a member of the hierarchy, generating a prompt comprising:
a path to the member, wherein the path comprises one or more ancestor nodes and a delimiter between the one or more ancestor nodes and the member;
one or more values identified as having the pattern or anomaly for the member;
one or more metrics that define an extent of the pattern or anomaly in relation to one or more other values;
one or more time periods during which the pattern or anomaly occurred;
one or more examples comprising one or more example paths, one or more example values, one or more example metrics, one or more example time periods, and one or more example summaries of one or more example patterns or anomalies for one or more example members at the one or more example paths;
prompting a large language model with the prompt to generate a resulting summary of the pattern or anomaly;
causing display of at least part of the resulting summary of the pattern or anomaly.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, wherein the one or more other values is a mean value and wherein the anomaly is a detected by a determination that the one or more metrics that define an extent of the pattern or anomaly in relation to one or more other values is three standard deviations from a mean value.

### Claim 3 (depends on 1)

The computer-implemented method of claim 1, wherein the one or more other values is a forecasted value generated by a mathematical trend of a set of data and wherein the one or more values identified as having the pattern or anomaly for the member is a recorded value.

### Claim 4 (depends on 1)

The computer-implemented method of claim 1, wherein the one or more other values is a forecasted value generated by a mathematical trend of a set of data and wherein the one or more values identified as having the pattern or anomaly for the member is a predicted value generated by a machine learning model trained to predict values for the hierarchy of data.

### Claim 5 (depends on 1)

The computer-implemented method of claim 1, wherein the one or more example paths comprises the delimiter between one or more example ancestors and the one or more example members.

### Claim 6 (depends on 1)

The computer-implemented method of claim 1, wherein the method further comprises:
receiving user feedback for the resulting summary indicating a modification to the resulting summary;
re-prompting the large language model to generate a modified summary based on the indicated modification;
causing display of at least part of the modified summary.

### Claim 7 (depends on 1)

The computer implemented method of claim 1, wherein the method is performed by an artificial intelligence agent trained on multi-dimensional training data of a same data type as the multi-dimensional set of data, wherein one or more other data types are handled by one or more other agents of a plurality of agents, and wherein the method further comprises:
filtering the data of the multi-dimensional set of data based on a method of filtering relevant to the data type of the multi-dimensional set of data.

### Claim 8

A computer-program product comprising one or more non-transitory machine-readable storage media, including stored instructions configured to cause a computing system to perform a set of actions including:
determining that analysis is enabled for a hierarchy of data of a multi-dimensional set of data, wherein analysis is disabled for one or more other hierarchies of data of the multi-dimensional set of data;
based at least in part on determining that the analysis is enabled, analyzing members of the hierarchy to detect a pattern or anomaly within the hierarchy;
in response to detecting the pattern or anomaly in a member of the hierarchy, generating a prompt comprising:
a path to the member, wherein the path comprises one or more ancestor nodes and a delimiter between the one or more ancestor nodes and the member;
one or more values identified as having the pattern or anomaly for the member;
one or more metrics that define an extent of the pattern or anomaly in relation to one or more other values;
one or more time periods during which the pattern or anomaly occurred;
one or more examples comprising one or more example paths, one or more example values, one or more example metrics, one or more example time periods, and one or more example summaries of one or more example patterns or anomalies for one or more example members at the one or more example paths;
prompting a large language model with the prompt to generate a resulting summary of the pattern or anomaly;
causing display of at least part of the resulting summary of the pattern or anomaly.

### Claim 9 (depends on 8)

The computer-program product of claim 8, wherein the one or more other values is a mean value and wherein the anomaly is a detected by a determination that the one or more metrics that define an extent of the pattern or anomaly in relation to one or more other values is three standard deviations from a mean value.

### Claim 10 (depends on 8)

The computer-program product of claim 8, wherein the one or more other values is a forecasted value generated by a mathematical trend of a set of data and wherein the one or more values identified as having the pattern or anomaly for the member is a recorded value.

### Claim 11 (depends on 8)

The computer-program product of claim 8, wherein the one or more other values is a forecasted value generated by a mathematical trend of a set of data and wherein the one or more values identified as having the pattern or anomaly for the member is a predicted value generated by a machine learning model trained to predict values for the hierarchy of data.

### Claim 12 (depends on 8)

The computer-program product of claim 8, wherein the one or more example paths comprises the delimiter between one or more example ancestors and the one or more example members.

### Claim 13 (depends on 8)

The computer-program product of claim 8, wherein the set of actions further includes:
receiving user feedback for the resulting summary indicating a modification to the resulting summary;
re-prompting the large language model to generate a modified summary based on the indicated modification;
causing display of at least part of the modified summary.

### Claim 14 (depends on 8)

The computer-program product of claim 8, wherein the set of actions are performed by an artificial intelligence agent trained on multi-dimensional training data of a same data type as the multi-dimensional set of data, wherein one or more other data types are handled by one or more other agents of a plurality of agents, and wherein the set of actions further includes:
filtering the data of the multi-dimensional set of data based on a method of filtering relevant to the data type of the multi-dimensional set of data.

### Claim 15

A system comprising:
one or more processors;
one or more non-transitory computer-readable media storing instructions, which, when executed by the system, cause the system to perform a set of actions including:
determining that analysis is enabled for a hierarchy of data of a multi-dimensional set of data, wherein analysis is disabled for one or more other hierarchies of data of the multi-dimensional set of data;
based at least in part on determining that the analysis is enabled, analyzing members of the hierarchy to detect a pattern or anomaly within the hierarchy;
in response to detecting the pattern or anomaly in a member of the hierarchy, generating a prompt comprising:
a path to the member, wherein the path comprises one or more ancestor nodes and a delimiter between the one or more ancestor nodes and the member;
one or more values identified as having the pattern or anomaly for the member;
one or more metrics that define an extent of the pattern or anomaly in relation to one or more other values;
one or more time periods during which the pattern or anomaly occurred;
one or more examples comprising one or more example paths, one or more example values, one or more example metrics, one or more example time periods, and one or more example summaries of one or more example patterns or anomalies for one or more example members at the one or more example paths;
prompting a large language model with the prompt to generate a resulting summary of the pattern or anomaly;
causing display of at least part of the resulting summary of the pattern or anomaly.

### Claim 16 (depends on 15)

The system of claim 15, wherein the one or more other values is a mean value and wherein the anomaly is a detected by a determination that the one or more metrics that define an extent of the pattern or anomaly in relation to one or more other values is three standard deviations from a mean value.

### Claim 17 (depends on 15)

The system of claim 15, wherein the one or more other values is a forecasted value generated by a mathematical trend of a set of data and wherein the one or more values identified as having the pattern or anomaly for the member is a recorded value.

### Claim 18 (depends on 15)

The system of claim 15, wherein the one or more other values is a forecasted value generated by a mathematical trend of a set of data and wherein the one or more values identified as having the pattern or anomaly for the member is a predicted value generated by a machine learning model trained to predict values for the hierarchy of data.

### Claim 19 (depends on 15)

The system of claim 15, wherein the one or more example paths comprises the delimiter between one or more example ancestors and the one or more example members.

### Claim 20 (depends on 15)

The system of claim 15, wherein the set of actions further includes:
receiving user feedback for the resulting summary indicating a modification to the resulting summary;
re-prompting the large language model to generate a modified summary based on the indicated modification;
causing display of at least part of the modified summary.
