Consumer decision tree generation system

• US 8,874,499 B2
• Filed: 06/21/2012
• Issued: 10/28/2014
• Est. Priority Date: 06/21/2012
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Wu, Su-Ming, Tiwari, Sandeep, Gaidarev, Peter
• Primary Examiner(s): Starks, Wilbert L
• Assistant Examiner(s): Wu, Fuming
• Application Number:
  US13/529,460
• Publication Number:
  US 20130346352A1
• Time in Patent Office: 859 Days
• Field of Search: None
• US Class Current: 706/46
• CPC Class Codes: G06Q 30/00 Commerce
• CPC Class Codes: G06Q 30/0278 Product appraisal
• CPC Class Codes: G06Q 50/10 Services

Abstract

A system generates a consumer decision tree (“CDT”). The system receives customer purchasing data that includes transactions of a plurality of products each having at least one product attribute. For a product category, the system identifies a plurality of similar products from the purchasing data and one or more attributes corresponding to each similar product. The system assigns the product category as a current level of the CDT, and determines a most significant attribute of the plurality of attributes for the current level. The system forms a next level of the CDT by dividing the most significant attribute into a plurality of sub-sections, where each sub-section corresponds to an attribute value of the most significant attribute. The system then forms a next level of the CDT for each sub-section until a terminal node is identified.

Claims

1. A non-transitory computer readable medium having instructions stored thereon that, when executed by a processor, causes the processor to generate a consumer decision tree (CDT) comprising a plurality of nodes, the generating of the consumer decision tree comprising:

receiving customer purchasing data comprising transactions of a plurality of products each having at least one product attribute;

for a product category, identifying a plurality of similar products from the purchasing data and one or more attributes corresponding to each similar product;

assigning the product category as a current level of the CDT;

determining a most significant attribute of the plurality of attributes for the current level, the determining the most significant attribute comprising determining a correlation between a dissimilarity between a product pair and a dissimilarity for each of the attributes, and then selecting the attribute having a dissimilarity with the highest correlation with the product dissimilarity;

forming a next level of the CDT by dividing the most significant attribute into a plurality of sub-sections, wherein each sub-section corresponds to an attribute value of the most significant attribute; and

assigning each of the sub-sections as the current level and repeating the determining the most significant attribute and forming the next level of the CDT for each sub-section until a terminal node is identified.

2. The computer readable medium of claim 1, wherein the customer purchasing data is generated by a loyalty program of a retailer.

3. The computer readable medium of claim 1, wherein identifying the plurality of similar products from the purchasing data having a plurality of customers comprises:

designating a first pair of products comprising a first product and a second product as similar when purchased by a first customer anytime during a transaction history of the first customer;

designating the first pair of products as dissimilar when purchased by a second customer in a same transaction;

for each of the customers, generating a factor comprising a proportion of transactions where the first pair of product out of all transactions where either the first product or the second product was purchased.

4. The computer readable medium of claim 1, wherein the customer purchasing data is generated by an enterprise resource planning system.

5. The computer readable medium of claim 1, further comprising:

identifying functional-fit attributes for the product category;

assigning the functional-fit attributes as the first level of the CDT directly below the product category.

6. The computer readable medium of claim 1, wherein the terminal node is identified when no significant attribute is identified.

7. The computer readable medium of claim 1, wherein the terminal node is identified when a number of items in a node is less than a predetermined percentage of the total items in the product category.

8. The computer readable medium of claim 1, wherein the terminal node is identified when an average dissimilarity of a child node is greater than a corresponding parent node.

9. The computer readable medium of claim 1, further comprising generating a score of the CDT.

10. A computer implemented method for generating a consumer decision tree (CDT) comprising a plurality of nodes, the method comprising:

receiving customer purchasing data comprising transactions of a plurality of products each having at least one product attribute;

for a product category, identifying a plurality of similar products from the purchasing data and one or more attributes corresponding to each similar product;

assigning the product category as a current level of the CDT;

determining a most significant attribute of the plurality of attributes for the current level, the determining the most significant attribute comprising determining a correlation between a dissimilarity between a product pair and a dissimilarity for each of the attributes, and then selecting the attribute having a dissimilarity with the highest correlation with the product dissimilarity;

forming a next level of the CDT by dividing the most significant attribute into a plurality of sub-sections, wherein each sub-section corresponds to an attribute value of the most significant attribute; and

assigning each of the sub-sections as the current level and repeating the determining the most significant attribute and forming the next level of the CDT for each sub-section until a terminal node is identified.

11. The method of claim 10, wherein identifying the plurality of similar products from the purchasing data having a plurality of customers comprises:

designating a first pair of products comprising a first product and a second product as similar when purchased by a first customer anytime during a transaction history of the first customer;

designating the first pair of products as dissimilar when purchased by a second customer in a same transaction; and

for each of the customers, generating a factor comprising a proportion of transactions where the first pair of product out of all transactions where either the first product or the second product was purchased.

12. The method of claim 10, wherein the terminal node is identified when no significant attribute is identified.

13. The method of claim 10, wherein the terminal node is identified when a number of items in a node is less than a predetermined percentage of the total items in the product category.

14. The method of claim 10, wherein the terminal node is identified when an average dissimilarity of a child node is greater than a corresponding parent node.

15. The method of claim 10, further comprising generating a score of the CDT.

16. A consumer decision tree (CDT) generation system comprising:

a processor;

a memory coupled to the processor and storing instructions that cause the processor to;

receive customer purchasing data comprising transactions of a plurality of products each having at least one product attribute;

for a product category, identify a plurality of similar products from the purchasing data and one or more attributes corresponding to each similar product;

assign the product category as a current level of the CDT;

determine a most significant attribute of the plurality of attributes for the current level, the determining the most significant attribute comprising determining a correlation between a dissimilarity between a product pair and a dissimilarity for each of the attributes, and then selecting the attribute having a dissimilarity with the highest correlation with the product dissimilarity;

form a next level of the CDT by dividing the most significant attribute into a plurality of sub-sections, wherein each sub-section corresponds to an attribute value of the most significant attribute; and

assign each of the sub-sections as the current level and repeating the determine the most significant attribute and form the next level of the CDT for each sub-section until a terminal node is identified.

17. The system of claim 16, wherein the identify the plurality of similar products from the purchasing data having a plurality of customers comprises:

designate a first pair of products comprising a first product and a second product as similar when purchased by a first customer anytime during a transaction history of the first customer;

designate the first pair of products as dissimilar when purchased by a second customer in a same transaction; and

for each of the customers, generate a factor comprising a proportion of transactions where the first pair of product out of all transactions where either the first product or the second product was purchased.

18. The system of claim 16, wherein the terminal node is identified when no significant attribute is identified.

19. The system of claim 16, wherein the terminal node is identified when a number of items in a node is less than a predetermined percentage of the total items in the product category.

20. The system of claim 16, wherein the terminal node is identified when an average dissimilarity of a child node is greater than a corresponding parent node.