Pair-wise graph querying, merging, and computing for account linking

Patent metadata

Field	Value
Publication number	US-2026080101-A1
Application number	US-202519367070-A
Country	US
Kind code	A1
Filing date	Oct 23, 2025
Priority date	Sep 12, 2022
Publication date	Mar 19, 2026
Grant date	—

Abstract

There are provided systems and methods for pairwise graph querying, merging, and computing for account linking. A service provider may provide an account graph system to identify pairwise similarities between different accounts based on shared data that may be identified through one or more linking characteristics. When providing pairwise graph similarities, a service provider may receive a query identifying two or more accounts and/or an account with a parameter for graph exploration and querying.
The service provider may utilize connection, link, or relationship graphs, queried and generated using a graph database, to determine pairwise similarities between the designated seed account and one or more selected accounts. The graph may include vertices for different queried data points and edges connecting such queries, where directionality of the edges or other vectors may be used to identify links or hops between accounts for data querying and exploration.

First claim

1 . (canceled)

2 . A service provider system comprising:
a non-transitory memory; and one or more hardware processors coupled to the non-transitory memory and configured to read instructions from the non-transitory memory to cause the service provider system to perform operations comprising:
determining a first graph for a first account based on a parameter associated with a plurality of first account activities for the first account, wherein the first graph comprises a plurality of first graph paths that associate the first account with a plurality of additional accounts based on the plurality of first account activities;
determining a second account linked to the first account based on at least one of the plurality of first graph paths, wherein the first account and the second account share at least one of the plurality of first account activities;
merging the first graph with a second graph associated with the second account, wherein the second graph comprises a plurality of second graph paths that associate the second account with at least one of the plurality of additional accounts based on a plurality of second account activities of the second account that are associated with the parameter;
identifying a similarity between the plurality of first graph paths and the plurality of second graph paths based on the merging; and
determining a risk of the first account using a risk analysis model based on the similarity between the plurality of first and second graph paths.

3 . The service provider system of claim 2 , wherein, prior to the determining the first graph, the operations further comprise:
receiving a request for a use of the first account, wherein the request identifies the first account and the parameter; and querying a graph database for the first graph based on the first account and the parameter.

4 . The service provider system of claim 3 , wherein the request is associated with one of an account promotional strategy, a risk review operation, a fraud detection operation, or an account compliance check.

5 . The service provider system of claim 2 , wherein the operations further comprise:
determining a subgraph of the merged first and second graphs based on at least a portion of at least one of the plurality of first graph paths and the plurality of second graph paths, wherein the risk analysis is further determined based on the subgraph.

6 . The service provider system of claim 2 , wherein the parameter comprises one of an IP address, payment card information, or information associated with the first account, and wherein one or more vertices of the first graph are associated with the parameter.

7 . The service provider system of claim 2 , wherein the identifying the similarity is further based on a pairwise similarity of shared vertices and a threshold number of the shared vertices.

8 . The service provider system of claim 2 , wherein the operations further comprise:
determining that one or more shared graph paths between the first account and the second account are less than or equal to a threshold path length; and generating a notification identifying that the second account is linked to the first account based on the determining that the one or more shared graph paths are less than or equal to the threshold path length.

9 . The service provider system of claim 2 , wherein the risk comprises a likelihood of risk collusion between the first account and the second account based on the similarity.

10 . A method comprising:
determining a first graph for a first account and a second graph for a second account based on one or more of a plurality of account activities by at least one of the first account or the second account; merging the first graph and the second graph based on a plurality of first graph paths for the first graph and a plurality of second graph paths for the second graph, wherein the merging identifies shared nodes in the plurality of first graph paths and the plurality of second graph paths; computing one or more similarities between the first graph and the second graph based on the merging and a traversal of the plurality of first and second graph paths; and determining a likelihood of risk collusion between the first graph and the second graph based on the computed one or more similarities.

11 . The method of claim 10 , wherein, prior to the determining the first graph, the method further comprises:
receiving a request for a use of the first account, wherein the request identifies the first account and a parameter associated with the plurality of account activities; and querying a graph database for the first graph based on the first account and the parameter.

12 . The method of claim 11 , wherein the request is associated with one of an account promotional strategy, a risk review operation, a fraud detection operation, or an account compliance check.

13 . The method of claim 10 , further comprising:
determining a subgraph of the merged first and second graphs based on an overlap between at least one of each of the plurality of first graph paths and the plurality of second graph paths, wherein the likelihood of risk collusion is determined further based on the subgraph.

14 . The method of claim 10 , wherein the first graph is further determined based on a parameter comprising one of an IP address, payment card information, or information associated with the first account, and wherein one or more vertices of the first graph are associated with the parameter.

15 . The method of claim 10 , wherein the one or more similarities is determined further based on a pairwise similarity of shared vertices and a threshold number of the shared vertices.

16 . The method of claim 10 , further comprising:
determining that a shared graph path between the first account and the second account of the merged first and second graphs is less than or equal to a threshold path length, wherein the likelihood of risk collusion is determined further based on the determining that the shared graph path is less than or equal to the threshold path length.

17 . The method of claim 16 , further comprising:
generating a notification identifying that the second account is associated with the first account based on the shared graph path.

18 . A non-transitory machine-readable medium having instructions stored thereon, the instructions executable to cause performance of operations comprising:
linking a first account to a second account based on at least one of a plurality of first graph paths of a first graph for the first account, wherein the plurality of first graph paths associate the first account with a plurality of additional accounts based on a plurality of first account activities associated with an account parameter, and wherein the first account and the second account share at least one of the plurality of first account activities; merging the first graph with a second graph for the second account, wherein the second graph comprises a plurality of second graph paths that associate the second account with at least one of the plurality of additional accounts; identifying one or more overlaps between the plurality of first graph paths and the plurality of second graph paths based on the merging; computing a similarity between the first account and the second account based on the one or more overlaps; and determining a risk associated with the first account using a risk analysis model and the computed similarity.

19 . The non-transitory machine-readable medium of claim 18 , wherei

Assignees

• Paypal Inc

Inventors

• Zhang Pengshan
• Wiener Alon
• Liu Delin
• Zhang Haoran
• Levi Itzik
• Guo Junshi
• Lin Ying
• Zhang Yu
• Marad Zohar Li

Classifications

• G06F21/6263 Primary

  during internet communication, e.g. revealing personal data from cookies · CPC title