360 degree framework

Patent metadata

Field	Value
Publication number	US-11915324-B2
Application number	US-201716621602-A
Country	US
Kind code	B2
Filing date	Jun 16, 2017
Priority date	Jun 16, 2017
Publication date	Feb 27, 2024
Grant date	Feb 27, 2024

Abstract

An architecture of connected servers supports data analysis with each server using a pattern matching algorithm to determine if an individual's traits match a predetermined species or if a new species should be established. Each server may have a dedicated database and receive information from relevant sources including various reporting agencies.

First claim

The invention claimed is:

1. A method of characterizing a user into a species, the method comprising:
receiving, by an input system of a server, activity data related to activity of the user, wherein the user is associated with user accounts, and wherein the activity data comprises transactions of the user accounts;
classifying, by a first analysis tool of the server, the activity data according to activity categories, wherein the activity categories include at least one of purchases, transfers, bill payments, borrowing activity, savings activity, or investment activity, or a combination thereof;
identifying, by the first analysis tool of the server, characteristic values based on the activity data, wherein each of the characteristic values corresponds to a different one of the activity categories;
identifying, by the first analysis tool of the server, a tracking characteristic value for at least one of the activity categories, wherein the tracking characteristic value is identified based on a frequency with which the user electronically accesses information related to at least one of the user accounts, and wherein the characteristic values and the tracking characteristic value define a set of user characteristic values;
comparing, by a second analysis tool of the server, the set of user characteristic values to sets of species characteristic values based on a distance function, wherein each of the sets of species characteristic values corresponds to a different known species, wherein the distance function generates a sum of distances for each of the known species, and wherein a closest match species corresponds to a lowest sum of distances;
comparing, by the second analysis tool of the server, the lowest sum of distances to a threshold value;
characterizing, by the second analysis tool of the server, the user into the closest match species when the lowest sum of distances is below the threshold value; and
characterizing, by the second analysis tool of the server, the user into a new species when the lowest sum of distances is above the threshold value.

2. The method of claim 1 , wherein identifying characteristic values based on the activity data comprises identifying a purchase characteristic.

3. The method of claim 2 , wherein identifying characteristic values based on the activity data comprises identifying a transfer characteristic.

4. The method of claim 3 , wherein identifying characteristic values based on the activity data comprises identifying a bill payment characteristic.

5. The method of claim 4 , wherein identifying characteristic values based on the activity data comprises identifying an investment activity characteristic.

6. The method of claim 5 , wherein identifying characteristic values based on the activity data comprises identifying a savings activity characteristic.

7. The method of claim 6 , wherein identifying characteristic values based on the activity data comprises identifying a borrowing activity characteristic.

8. The method of claim 1 , wherein identifying characteristic values based on the activity data comprises applying, by the second analysis tool of the server, the activity data to a machine learning tool, and wherein the machine learning tool includes a learning model that learns based on prior manual identification of characteristic values.

9. The method of claim 8 , wherein the learning model that changes over time.

10. A system for characterizing an individual into a species, comprising:
a plurality of servers dedicated to receiving activity data corresponding to activity of the individual, wherein the activity data comprises transactions of accounts associated with the individual, wherein each of the plurality of servers corresponds to a different activity category, and wherein the activity categories include at least one of purchases, transfers, bill payments, borrowing activity, savings activity, or investment activity, or a combination thereof;
at least one tracking server dedicated to receiving tracking data associated with how often the individual electronically accesses information related to the accounts;
a plurality of input systems, wherein each of the input systems corresponds to a respective one of the plurality of servers and the at least one tracking server, wherein each of the input systems collects the activity data or the tracking data;
a plurality of database systems, wherein each of the database systems corresponds to a respective one of the plurality of servers and the at least one tracking server, wherein each of the database systems stores the activity data or the tracking data;
a first analysis tool configured to determine a set of individual placement values for the individual, wherein the set of individual placement values comprises activity placement values corresponding to each of the activity categories and a tracking placement value corresponding to how often the individual electronically accesses information related to the accounts;
a second analysis tool configured to:
compare the set of individual placement values against existing sets of placement values for known species based on a distance function, wherein the distance function generates a sum of placement distances for each of the known species, and wherein a closest match species of the known species corresponds to a lowest sum of distances, and
determine whether the lowest sum of distances satisfies a threshold value;
responsive to the lowest sum of distances satisfying the threshold value characterize the individual into the closest match species; and
responsive to the lowest sum of distances failing to satisfy the threshold value characterize the individual as a new species.

11. The system of claim 10 , wherein the second analysis tool includes a module that generates a curve based on the set of individual placement values.

12. The system of claim 11 , wherein the curve is a first curve, wherein the module generates second curves based on the existing sets of placement values, and wherein the module performs a correlation between the first curve and each of the second curves.

Assignees

• Visa Int Service Ass

Inventors

• Chopra Akshay
• Dill Matthew

Classifications

• G06Q40/12 Primary

  Accounting · CPC title

• G06Q30/04

  Billing or invoicing · CPC title

• G06Q40/02

  Banking, e.g. interest calculation or account maintenance (credit or loans G06Q40/03) · CPC title