Customer clustering using integer programming

Patent metadata

Field	Value
Publication number	US-11823218-B2
Application number	US-202217705483-A
Country	US
Kind code	B2
Filing date	Mar 28, 2022
Priority date	Nov 20, 2013
Publication date	Nov 21, 2023
Grant date	Nov 21, 2023

Abstract

Methods and apparatus are disclosed regarding an e-commerce system that clusters customers based on demographic data and purchase history data for the customers. In some embodiments, the e-commerce system solves an Integer Program that accounts for the demographic data and purchase history data in order to identify a hyperplane that splits a selected cluster of customers.

First claim

What is claimed is:

1. A method comprising:
using one or more processors for:
tailoring a service for a particular customer according to a customer cluster that comprise the particular customer, wherein the customer cluster is one of a plurality of customer clusters; and
periodically updating the plurality of customer clusters to maximize inner-similarities among customers in each of the plurality of customer clusters,
wherein the inner-similarities are determined according to purchase history data and demographic data; and

using a classifier for:
solving an Integer Program that accounts for the purchase history data and the demographic data of a selected customer cluster; and
iteratively dividing customer sets into two partitions until a suitable number of partitions for a customer base is obtained.

2. The method of claim 1 , wherein the service comprises providing product recommendations according to the customer cluster comprising the particular customer.

3. The method of claim 1 , wherein the service comprises providing product promotions according to the customer cluster comprising the particular customer.

4. The method of claim 1 , wherein the service comprises providing coupons according to the customer cluster comprising the particular customer.

5. The method of claim 1 , wherein the inner-similarities are maximized according to one or more solutions of an integer program.

6. The method of claim 1 , wherein updating the plurality of customer clusters comprises selecting a particular customer cluster that has a population greater than a specified limit and splitting the particular customer cluster.

7. The method of claim 1 , wherein the method comprises storing the purchase history data in one or more relational database tables such that each row includes transaction data and a customer identifier that identifies a customer associated with the transaction data.

8. The method of claim 1 , wherein updating the plurality of customer clusters comprises coalescing purchased items of multiple item identifiers under a single identifier and updating the plurality of customer clusters according to the purchased items under the single identifier.

9. The method of claim 1 , wherein the plurality of customer clusters are updated according to a customer-item (CI) matrix, wherein:
each row of the CI matrix corresponds to a customer identifier,
each column of the CI matrix corresponds to a category identifier, and
each entry of the CI matrix corresponds to a quantity associated with one or more customer identifiers.

10. The method of claim 1 , wherein the plurality of customer clusters are updated according to a customer-item (CI) matrix, and wherein each column of the CI matrix is standardized according to a bin quantiles standardization (BQS).

11. A system comprising:
one or more processors configured to:
tailor a service for a particular customer according to a customer cluster that comprise the particular customer, wherein the customer cluster is one of a plurality of customer clusters; and
periodically update the plurality of customer clusters to maximize inner-similarities among customers in each of the plurality of customer clusters, wherein the inner-similarities are determined according to purchase history data and demographic data; and

a classifier configured to:
solve an Integer Program that accounts for the purchase history data and the demographic data of a selected cluster; and
iteratively divide customer sets into two partitions until a suitable number of partitions for a customer base is obtained.

12. The system of claim 11 , wherein the one or more processors are configured to provide product recommendations according to the customer cluster comprising the particular customer.

13. The system of claim 11 , wherein the one or more processors are configured to provide product promotions according to the customer cluster comprising the particular customer.

14. The system of claim 11 , wherein the one or more processors are configured to provide coupons according to the customer cluster comprising the particular customer.

15. The system of claim 11 , wherein the one or more processors are configured to maximize the inner-similarities according to one or more solutions of an integer program.

16. The system of claim 11 , wherein the one or more processors are configured to update the plurality of customer clusters by selecting a particular customer cluster that has a population greater than a specified limit and splitting the particular customer cluster.

17. The system of claim 11 , wherein the system comprises one or more relational database tables for storing the purchase history data such that each row includes transaction data and a customer identifier that identifies a customer associated with the transaction data.

18. The system of claim 11 , wherein the one or more processors are configured to coalesce purchased items of multiple item identifiers under a single identifier, and wherein the one or more processors are configured to update the plurality of customer clusters according to the purchased items under the single identifier.

19. The system of claim 11 , wherein the one or more processors are configured to update the plurality of customer clusters according to a customer-item (CI) matrix, wherein:
each row of the CI matrix corresponds to a customer identifier,
each column of the CI matrix corresponds to a category identifier, and
each entry of the CI matrix corresponds to a quantity associated with one or more customer identifiers.

20. The system of claim 11 , wherein the one or more processors are configured to update the plurality of customer clusters according to a customer-item (CI) matrix, and wherein each column of the CI matrix is standardized according to a bin quantiles standardization (BQS).

Assignees

• Transf Sr Brands Llc

Inventors

• Aydin Burcu
• Tamir Michael

Classifications

• G06Q30/0204 Primary

  Market segmentation · CPC title

• G06Q30/0251

  Targeted advertisements · CPC title