Back to results
US 11,636,486 B2Granted
Determining subsets of accounts using a model of transactions
AssigneePaypal, Inc.
InventorZhe Chen
Life of this patent
Estimating the term…
Abstract
A computer system receives a first set of information that describes a set of transactions between pairs of user accounts of a service and a second set of information that specifies attribute values for user accounts of the service that are recipient user accounts within the set of transactions. Using the first set of information and the second set of information, the computer system generates an augmented graph model of the transactions and attribute values. By applying modularity maximization to the augmented graph model, the computer system determines a plurality of subsets of recipient user accounts.
Claims20
A computer system comprising:
- a computer processor circuit;
- and
- a computer-memory storing instructions executable by the computer processor circuit to cause the computer system to perform operations including:
- receiving, at the computer system, a first set of information that describes a set of transactions between pairs of user accounts of a computer-implemented service, wherein a pair of user accounts for a given transaction includes an initiator user account and a recipient user account;
- receiving, at the computer system, a second set of information that specifies a set of attribute values for user accounts of the service that are recipient user accounts within the set of transactions;
- generating, by the computer system using the first set of information, a graph model representing user accounts as nodes and the set of transactions as edges between pairs of nodes, wherein the graph model includes at least one million nodes;
- identifying, with the computer system, a plurality of attribute clusters within the set of attribute values specified by the second set of information, wherein the attribute clusters include attributed nodes that have attribute values specified by the second set of information;
- creating an augmented graph model by inserting, for a given attribute cluster among a group of the attribute clusters, an additional node into the graph model and connecting, with weighted attribute edges, the additional node to attributed nodes in the given attribute cluster, wherein the additional node represents the attribute values of the attributed nodes clustered in the given attribute cluster;
- determining, with the computer system using topological information of the augmented graph model having the additional nodes corresponding to the attribute clusters in the group, a plurality of subsets of recipient user accounts;
- and
- managing, by the computer system, an aspect of the computer-implemented service using the determined plurality of subsets of recipient user accounts.
8 more elements
A method comprising:
- receiving, at a computer system, a first set of information that describes a set of transactions between pairs of user accounts of a computer-implemented service, wherein a pair of user accounts for a given transaction includes an initiator user account and a recipient user account;
- receiving, at the computer system, a second set of information that specifies attribute values for user accounts of the service that are recipient user accounts within the set of transactions;
- generating, by the computer system and using the first set of information and second set of information, an augmented graph model of the set of transactions and attribute values, wherein the augmented graph model includes at least one million nodes, wherein the augmented graph model has been augmented with respective additional nodes inserted into an original graph model, wherein a given additional node corresponds to a respective attribute cluster and is connected to attributed nodes in the respective attribute cluster, wherein a given additional node represents the attribute values of the attributed nodes clustered in a given attribute cluster;
- and
- determining, with the computer system by applying modularity maximization to the augmented graph model, a plurality of subsets of recipient user accounts, wherein the determining is based on topological information of the augmented graph model and the additional nodes corresponding to the respective attribute clusters.
2 more elements
A non-transitory, computer-readable medium storing instructions that when executed by a computer system cause the computer system to perform operations comprising:
- generating an augmented graph model representative of transactions between pairs of user accounts and attribute information about an attributed set of user accounts, wherein the attributed set of user accounts are represented in the augmented graph model as attributed nodes, wherein the augmented graph model has at least one million nodes and has been augmented with respective additional nodes inserted into an original graph model, wherein a given additional node corresponds to a respective attribute cluster and is connected to attributed nodes in the respective attribute cluster, and wherein a given additional node represents the attribute information of attributed nodes clustered in a given attribute cluster and is connected to respective attributed nodes in the given attribute cluster by respective attribute edges;
- determining, using modularity maximization applied to the attributed nodes in a sequence, a first grouping of the attributed nodes into subsets of user accounts, wherein the determining includes adjusting weights of the attribute edges of the first grouping;
- and
- determining, using modularity maximization applied to the attributed nodes in the same sequence, a second grouping of the attributed nodes into revised subsets of user accounts based on the first grouping and the adjusted weights of the attribute edges, wherein determining the second grouping for a given attributed node includes masking attributed nodes first grouped in the same subset as the given attributed node later in the sequence.
1 more element
Dependencies are inferred from the claim text.