# US 2025/0168075 A1 · Relational network inferencing using random traversal paths of a graph representation

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

- Number: US 2025/0168075 A1
- Document ID: US20250168075A1
- Status: Published application
- Kind: A1
- Application number: 18/515,605
- Filed: 2023-11-21
- Published: 2025-05-22
- Earliest claimed: 2023-11-21
- Assignees: Intuit Inc.
- Applicants: Intuit Inc.
- Inventors: Navid IMANI HOSSEIN ABAD; Nazanin Zaker HABIBABADI; Xue HAN
- Web page: https://patentgrep.com/patent/US20250168075A1

## CPC

- G06F 16/9024: {Graphs; Linked lists (G06F16/9027 takes precedence)}
- H04L 41/12: Discovery or management of network topologies
- G06Q 40/03: Credit; Loans; Processing thereof
- H04L 45/122: by minimising distances, e.g. by selecting a route with minimum of number of hops
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)
- 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

Certain aspects of the present disclosure provide techniques for performing inferencing for entities in a relational network by determining graph features of a graph representation of the relational network. Random paths are generated using each node of the graph representation as a starting node and using relative outgoing edge weights to determine next nodes along the traversal path. Probability values that nodes will be reached as secondary nodes along a traversal path are aggregated to generate a score value or graph feature for each node of the graph representation. The graph features may be used as training data to train a machine learning model to infer risk values or graph features based on additional transaction data input.

## Claims

### Claim 1

A method of resource-efficient relational network inferencing, comprising:
defining, for a graph representation, a plurality of traversal paths including:
a starting vertex having an initial probability value; and
a random hop selected based on a weight of an edge and a total weight of selectable edges for the random hop; and
determining a probability that an entity represented by a particular vertex will be impacted by an event associated with the starting vertex based on the initial probability value and a proportion of the plurality of traversal paths that includes the particular vertex.

### Claim 2 (depends on 1)

The method of claim 1, wherein:
the random hop is a first random hop selected based a first parameter of the weight; and
the plurality of traversal paths includes a second random hop selected based on a second parameter.

### Claim 3 (depends on 1)

The method of claim 1, wherein the plurality of traversal paths comprises a first traversal path having a first number of hops and a second traversal path having a second number of hops.

### Claim 4 (depends on 3)

The method of claim 3, wherein the plurality of traversal paths is defined using a minimum number of hops per path and a maximum number of hops per path.

### Claim 5 (depends on 1)

The method of claim 1, wherein:
the starting vertex is a first starting vertex;
the initial probability value is a first initial probability value;
the plurality of traversal paths includes a second starting vertex having a second initial probability value; and
the method comprises determining an aggregate probability that the entity represented by the particular vertex will be impacted by an event associated with at least one of the first starting vertex and the second starting vertex based on the first initial probability value, the second initial probability value, and the proportion of the plurality of traversal paths that includes the particular vertex.

### Claim 6 (depends on 1)

The method of claim 1, wherein the weight is determined by a template defining edge weight based on a selection from: a transaction type of an occurrence, a number of transaction occurrences, and an amount of a transaction payment.

### Claim 7 (depends on 1)

The method of claim 1, wherein:
the particular vertex represents a first entity, the starting vertex represents a second entity, the edge represents a transaction, the event comprises a default of the second entity; and
the method comprises determining a risk of default for the first entity based on the default of the second entity and the transaction.

### Claim 8 (depends on 1)

The method of claim 1, wherein:
the plurality of traversal paths is a first plurality of traversal paths; and
the method further comprises:
defining, for the graph representation, a second plurality of traversal paths starting at the particular vertex and using the probability that the entity represented by the particular vertex will be impacted; and
determining a second probability that a second entity represented by a second vertex will be impacted by the event associated with the starting vertex based on a proportion of the second plurality of traversal paths that includes the second vertex.

### Claim 9

A system comprising a computing device having a processor and a memory having executable instructions stored thereon, which, when executed, cause the processor to perform a method of:
defining, for a graph representation, a plurality of traversal paths including:
a starting vertex having an initial probability value; and
a random hop selected based on a weight of an edge and a total weight of selectable edges for the random hop; and
determining a probability that an entity represented by a particular vertex will be impacted by an event associated with the starting vertex based on the initial probability value and a proportion of the plurality of traversal paths that includes the particular vertex.

### Claim 10 (depends on 9)

The system of claim 9, wherein:
the random hop is a first random hop selected based a first parameter of the weight; and
the plurality of traversal paths includes a second random hop selected based on a second parameter.

### Claim 11 (depends on 9)

The system of claim 9, wherein the plurality of traversal paths comprises a first traversal path having a first number of hops and a second traversal path having a second number of hops.

### Claim 12 (depends on 11)

The system of claim 11, wherein the plurality of traversal paths is defined using a minimum number of hops per path and a maximum number of hops per path.

### Claim 13 (depends on 9)

The system of claim 9, wherein:
the starting vertex is a first starting vertex;
the initial probability value is a first initial probability value;
the plurality of traversal paths includes a second starting vertex having a second initial probability value; and
the method comprises determining an aggregate probability that the entity represented by the particular vertex will be impacted by an event associated with at least one of the first starting vertex and the second starting vertex based on the first initial probability value, the second initial probability value, and the proportion of the plurality of traversal paths that includes the particular vertex.

### Claim 14 (depends on 9)

The system of claim 9, wherein the weight is determined by a template defining edge weight based on a selection from: a transaction type of an occurrence, a number of transaction occurrences, and an amount of a transaction payment.

### Claim 15 (depends on 9)

The system of claim 9, wherein:
the particular vertex represents a first entity, the starting vertex represents a second entity, the edge represents a transaction, the event comprises a default of the second entity; and
the method comprises determining a risk of default for the first entity based on the default of the second entity and the transaction.

### Claim 16 (depends on 9)

The system of claim 9, wherein:
the plurality of traversal paths is a first plurality of traversal paths; and
the method further comprises:
defining, for the graph representation, a second plurality of traversal paths starting at the particular vertex and using the probability that the entity represented by the particular vertex will be impacted; and
determining a second probability that a second entity represented by a second vertex will be impacted by the event associated with the starting vertex based on a proportion of the second plurality of traversal paths that includes the second vertex.

### Claim 17

A non-transitory computer-readable medium comprising instructions that, when executed by a processor, cause the processor to perform a method for determining a graph feature, the method comprising:
defining, for a graph representation, a plurality of traversal paths including:
a starting vertex having an initial probability value; and
a random hop selected based on a weight of an edge and a total weight of selectable edges for the random hop; and
determining a probability that an entity represented by a particular vertex will be impacted by an event associated with the starting vertex based on the initial probability value and a proportion of the plurality of traversal paths that includes the particular vertex.

### Claim 18 (depends on 17)

The non-transitory computer-readable medium of claim 17, wherein the plurality of traversal paths comprises a first traversal path having a first number of hops and a second traversal path having a second number of hops; and the plurality of traversal paths is defined using a minimum number of hops per path and a maximum number of hops per path.

### Claim 19 (depends on 17)

The non-transitory computer-readable medium of claim 17, wherein:
the starting vertex is a first starting vertex;
the initial probability value is a first initial probability value;
the plurality of traversal paths includes a second starting vertex having a second initial probability value; and
the method comprises determining an aggregate probability that the entity represented by the particular vertex will be impacted by an event associated with at least one of the first starting vertex and the second starting vertex based on the first initial probability value, the second initial probability value, and the proportion of the plurality of traversal paths that includes the particular vertex.

### Claim 20 (depends on 17)

The non-transitory computer-readable medium of claim 17, wherein:
the plurality of traversal paths is a first plurality of traversal paths; and
the method further comprises:
defining, for the graph representation, a second plurality of traversal paths starting at the particular vertex and using the probability that the entity represented by the particular vertex will be impacted; and
determining a second probability that a second entity represented by a second vertex will be impacted by the event associated with the starting vertex based on a proportion of the second plurality of traversal paths that includes the second vertex.
