Graph learning and automated behavior coordination platform

Patent metadata

Field	Value
Publication number	US-12566809-B2
Application number	US-201917765774-A
Country	US
Kind code	B2
Filing date	Oct 1, 2019
Priority date	Oct 1, 2019
Publication date	Mar 3, 2026
Grant date	Mar 3, 2026

Abstract

A system receives transaction data from payment devices of a user. The system generates a transaction profile of the user based on the received transaction data. Based on the transactional parameters, the system determines a community of the user. The system further receives one or more predefined sub-groups of the community from a database. The one or more sub-groups define subsets of activities, geographic locations, and times. The system identifies one or more rules setting limits on the transactional parameters. Absent receiving additional user interactions, the system generates a suggested community for the user as a function of the one or more rules, the activities, geographic locations, and times of the one or more sub-groups of the community. The system generates a suggested activity to the user based on the suggested community.

First claim

What is claimed is:

1 . A computer-implemented method comprising:
receiving transaction data collected from a payment device of a first user; generating a first historic transaction profile of the first user based on the received transaction data, wherein the first historic transaction profile comprises first geographic locations, first times, and first activities; receiving second historic transaction profiles for second users, wherein each of the second historic transaction profiles comprises second geographic locations, second times, and second activities; determining a community of the first user based on the first historic transaction profile and second historic transaction profiles, wherein the community comprises the first user and the second users, wherein determining the community comprises constructing a graph structure representing the community, said graph structure comprising at least two vertices and an edge between the at least two vertices, said each of the two vertices representing an activity, and said edge represent a functional measure between activities connected by the edge; generating a coordination sub-group graph for the community from the graph structure, wherein the coordination sub-group graph comprises the first user and a subset of the second users, and wherein the coordination sub-group graph defines a subset of activities, geographic locations, and times common to the first historic transaction profile and a subset of the second historic transaction profiles; and generating a coordinated activity recommendation for the first user and at least one of the second users based on the coordination sub-group graph, wherein the coordinated activity recommendation comprises the first user and at least one of the second users coordinating to repeat and perform together at least one of the first activities of the first historic transaction profile and at least one of the second activities of the second historic transaction profiles, wherein the first historic transaction profile and the second historic transaction profiles indicate that the first user and the at least one of the second users previously performed the at least one of the first activities and the at least one of the second activities of the coordinated activity recommendation separately.

2 . The computer-implemented method of claim 1 , wherein the functional measure comprises a minimum weighted edit distance similarity measure based on times and geographic locations.

3 . The computer-implemented method of claim 1 , wherein the receiving transaction data comprises data collected from the payment device of the first user, including at least data generated from sensors or components of a computing device of the first user, usage data of the computing device, and usage data of software applications of the computing device.

4 . The computer-implemented method of claim 3 , wherein generating the first historic transaction profile of the first user comprises generating the first historic transaction profile based on the at least data generated from sensors or components of computing device of the user, the usage data of the computing device, and the usage data of software applications of the computing device.

5 . A system comprising:
a payment processing network server comprising a memory and a processor, wherein the memory comprises instructions executable by the processor for: receiving stored transaction data from one or more payment devices of a first user; generating a first historic transaction profile of the first user based on the received transaction data, wherein the first historic transaction profile comprises first transactional parameters and first device-based parameters; receiving second historic transaction profiles for second users, wherein each of the second historic transaction profiles comprises second transactional parameters and second device-based parameters;
determining a community of the first user based on the first historic transaction profile and second historic transaction profiles, wherein the community comprises the first user and the second users, wherein determining the community comprises constructing a graph structure representing the community, said graph structure comprising at least two vertices and an edge between the at least two vertices, said each of the two vertices representing an activity, and said edge represent a functional measure between activities connected by the edge;
generating a coordination sub-group graph for the community from the graph structure, wherein the coordination sub-group graph comprises the first user and a subset of the second users, and wherein the coordination sub-group graph defines a subset transactional parameters and device-based parameters common to the first historic transaction profile and a subset of the second historic transaction profiles; generating a coordinated activity recommendation for the first user and at least one of the second users based on the coordination sub-group graph, wherein the coordinated activity recommendation comprises the first user and at least one of the second users coordinating to repeat and perform together a first activity corresponding to the first historic transaction profile and a second activity corresponding to the second historic transaction profile, wherein the first historic transaction profile and the second historic transaction profile indicate that the first user and the at least one of the second users previously performed the first activity and second activity separately.

6 . The system of claim 5 , wherein the transactional parameters common to the first historic transaction profile and the subset of the second historic transaction profiles comprise at least a geographic location parameter, a time parameter, and an activity parameter.

7 . The system of claim 5 , wherein the device-based parameters common to the first historic transaction profile and the subset of the second historic transaction profiles comprise at least data generated from sensors or components of the one or more payment devices of the first user, usage data of the one or more payment device of the first user, and usage data of software applications of the one or more payment devices of the first user.

8 . The system of claim 7 , wherein the one or more of the payment devices of the first user comprise a portable computing device.

9 . The system of claim 6 , wherein the functional measure comprises a minimum weighted edit distance similarity measure based on the time parameters and the geographic location parameters.

10 . The system of claim 8 , the memory further comprises instructions executable by the processor for collecting data corresponding to the first device-based parameters from the one or more payment devices of the first user.

11 . A non-transitory computer-readable medium stored thereon computer-executable instructions for:
receiving stored transaction data from one or more payment devices of a first user; generating a first historic transaction profile of the first user based on the received transaction data, wherein the first historic transaction profile comprises a first set of parameters corresponding to first activities; receiving second historic transaction profiles for second users, wherein each of the second historic transaction profiles comprises a second set of parameters corresponding to second activities; determining a community of the first user based on the first historic transaction profile and second historic transaction profiles, wherein the community comprises the first user and the second users, wherein determining the community comprises constructing a graph structure repre

Assignees

• Visa Int Service Ass

Inventors

• Harris Theodore
• Korolevskaya Tatiana
• Li Yue
• Wald Jerry

Classifications

• G06Q10/48

  using social graphs · CPC title

• G06Q10/42

  Determination of affinities or common interests between users · CPC title

• G06Q50/40 Primary

  Business processes related to the transportation industry (shipping G06Q10/083) · CPC title

• G06Q20/40

  Authorisation, e.g. identification of payer or payee, verification of customer or shop credentials; Review and approval of payers, e.g. check credit lines or negative lists · CPC title

• G06F16/29

  Geographical information databases · CPC title