# US 2021/0279279 A1 · Automated graph embedding recommendations based on extracted graph features

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

- Number: US 2021/0279279 A1
- Document ID: US20210279279A1
- Status: Published application
- Kind: A1
- Application number: 16/809,679
- Filed: 2020-03-05
- Published: 2021-09-09
- Earliest claimed: 2020-03-05
- Assignees: International Business Machines Corporation
- Applicants: International Business Machines Corporation
- Inventors: Ana Paula Appel; Renato Luiz de Freitas Cunha; Bruno Silva; Rogerio Abreu de Paula
- Web page: https://patentgrep.com/patent/US20210279279A1

## CPC

- G06F 16/9024: {Graphs; Linked lists (G06F16/9027 takes precedence)}
- G06F 18/213: Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F 18/22: Matching criteria, e.g. proximity measures
- G06F 18/2323: based on graph theory, e.g. minimum spanning trees [MST] or graph cuts
- G06N 3/08: Learning methods
- G06N 20/00: Machine learning
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

Embodiments of the invention are directed to a computer-implemented method for matching a graph-under-analysis to a technique for embedding the graph-under-analysis. In a non-limiting example, the computer-implemented method includes receiving, using a processor, graph data representing the graph-under-analysis, wherein the graph-under-analysis represents a network. The graph data is analyzed, using the processor, to extract graph property data representing properties of the graph-under-analysis. Based at least in part on a result of analyzing the graph property data, one or more embedding techniques are selected, wherein at least one of the one or more embedding techniques is configured to transform the graph data to a graph embedding that is used by a task algorithm to perform a task.

## Claims

### Claim 1

A computer-implemented method for matching a graph-under-analysis to a technique for embedding the graph-under-analysis, the computer-implemented method comprising:
receiving, using a processor, graph data representing the graph-under-analysis, wherein the graph-under-analysis represents a network;
analyzing, using the processor, the graph data to extract graph property data representing properties of the graph-under-analysis; and
based at least in part on a result of analyzing the graph property data, selecting one or more embedding techniques;
wherein at least one of the one or more embedding techniques is configured to transform the graph data to a graph embedding that is used by a task algorithm to perform a task.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, wherein:
analyzing the graph data comprises computing spectrum data of the graph-under-analysis; and
at least some of the graph property data is derived from the spectrum data of the graph-under-analysis.

### Claim 3 (depends on 1)

The computer-implemented method of claim 1, wherein analyzing the graph property data comprises applying a set of embedding technique selection rules to the graph property data.

### Claim 4 (depends on 3)

The computer-implemented method of claim 3, wherein:
the set of embedding technique selection rules includes comparing the graph property data to stored graph property data;
the stored graph property data is associated with stored graph embedding techniques; and
selecting the one or more embedding techniques is further based at least in part on a result of using the set of embedding technique selection rules to compare the graph property data to the stored graph property data.

### Claim 5 (depends on 1)

The computer-implemented method of claim 1, wherein analyzing the graph property data comprises applying the graph property data to a machine learning model.

### Claim 6 (depends on 5)

The computer-implemented method of claim 5, wherein the machine learning model determines a level of similarity between the graph property data and stored graph property data.

### Claim 7 (depends on 1)

The computer-implemented method of claim 1 further comprising:
analyzing the graph property data to determine a set of task options;
receiving a selected task, wherein the selected task is one of the task options;
receiving parameters of the selected task; and
receiving an instruction to apply a first one of the one or more embedding techniques to the graph data.

### Claim 8

A computer system for matching a graph-under-analysis to a technique for embedding the graph-under-analysis, the computer system comprising:
a memory; and
a processor communicatively coupled to the memory and configured to perform operations comprising:
receiving graph data representing the graph-under-analysis, wherein the graph-under-analysis represents a network;
analyzing the graph data to extract graph property data representing properties of the graph-under-analysis; and
based at least in part on a result of analyzing the graph property data, selecting one or more embedding techniques;
wherein at least one of the one or more embedding techniques is configured to transform the graph data to a graph embedding that is used by a task algorithm to perform a task.

### Claim 9 (depends on 8)

The computer system of claim 8, wherein:
analyzing the graph data comprises computing spectrum data of the graph-under-analysis; and
at least some of the graph property data is derived from the spectrum data of the graph-under-analysis.

### Claim 10 (depends on 8)

The computer system of claim 8, wherein analyzing the graph property data comprises applying a set of embedding technique selection rules to the graph property data.

### Claim 11 (depends on 10)

The computer system of claim 10, wherein:
the set of embedding technique selection rules includes comparing the graph property data to stored graph property data;
the stored graph property data is associated with stored graph embedding techniques; and
selecting the one or more embedding techniques is further based at least in part on a result of using the set of embedding technique selection rules to compare the graph property data to the stored graph property data.

### Claim 12 (depends on 8)

The computer system of claim 8, wherein analyzing the graph property data comprises applying the graph property data to a machine learning model.

### Claim 13 (depends on 12)

The computer system of claim 12, wherein the machine learning model determines a level of similarity between the graph property data and stored graph property data.

### Claim 14 (depends on 8)

The computer system of claim 8 further comprising:
analyzing the graph property data to determine a set of task options;
receiving a selected task, wherein the selected task is one of the task options;
receiving parameters of the selected task; and
receiving an instruction to apply a first one of the one or more embedding techniques to the graph data.

### Claim 15

A computer program product for matching a graph-under-analysis to a technique for embedding the graph-under-analysis, the computer program product comprising a computer readable program stored on a computer readable storage medium, wherein the computer readable program, when executed on a processor system, causes the processor system to perform operations comprising:
receiving graph data representing the graph-under-analysis, wherein the graph-under-analysis represents a network;
analyzing the graph data to extract graph property data representing properties of the graph-under-analysis;
wherein analyzing the graph data comprises computing spectrum data of the graph-under-analysis;
wherein at least some of the graph property data is derived from the spectrum data of the graph-under-analysis; and
based at least in part on a result of analyzing the graph property data, selecting one or more embedding techniques;
wherein at least one of the one or more embedding techniques is configured to transform the graph data to a graph embedding that is used by a task algorithm to perform a task.

### Claim 16 (depends on 15)

The computer program product of claim 15, wherein analyzing the graph property data comprises applying a set of embedding technique selection rules to the graph property data.

### Claim 17 (depends on 16)

The computer program product of claim 16, wherein:
the set of embedding technique selection rules includes comparing the graph property data to stored graph property data;
the stored graph property data is associated with stored graph embedding techniques; and
selecting the one or more embedding techniques is further based at least in part on a result of using the set of embedding technique selection rules to compare the graph property data to the stored graph property data.

### Claim 18 (depends on 15)

The computer program product of claim 15, wherein analyzing the graph property data comprises applying the graph property data to a machine learning model.

### Claim 19 (depends on 15)

The computer program product of claim 15, wherein the machine learning model determines a level of similarity between the graph property data and stored graph property data.

### Claim 20 (depends on 15)

The computer program product of claim 15, wherein the operations performed by the processor system further comprise:
analyzing the graph property data to determine a set of task options;
receiving a selected task, wherein the selected task is one of the task options;
receiving parameters of the selected task; and
receiving an instruction to apply a first one of the one or more embedding techniques to the graph data.
