SYSTEMS AND METHODS FOR MAKING CONTEXTUAL RECOMMENDATIONS

• US 20110010324A1
• Filed: 10/21/2009
• Published: 01/13/2011
• Est. Priority Date: 07/08/2009
• Status: Active Grant

• Current Assignee: eBay Inc.
• Original Assignee: eBay Inc.
• Inventors: Zheng, Jiaqian, Wu, Xiaoyuan, Bolivar, Alvaro
• Granted Patent: US 8,386,406 B2
• Time in Patent Office: Days
• US Class Current: 706/46
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G06Q 30/0255 based on user history
• CPC Class Codes: G06Q 30/0631 Item recommendations
• CPC Class Codes: H04L 67/306 User profiles
• CPC Class Codes: H04L 67/535 Tracking the activity of the user network monitoring arrangements H04L43/00; recording of computer activity G06F11/34

Abstract

A method and a system for making contextual recommendations to users on a network-based system. For example, a first and second related category are selected and a relationship score is calculated representing an aspect of the relationship between the first and second related categories. A relationship type is selected to represent the relationship between the first and second categories using the relationship score. The relationship type is one of either a complement or a substitute. A complement relationship indicates a complementary relationship between the first and second related categories. While a substitute relationship indicates a substitutionary relationship between the first and second categories. The relationship type is stored associated with the first and second related categories. Using the selected relationship type, a recommendation is provided to a user on the network-based system.

Claims

1. A method of creating relationships within a network-based system for making recommendations, the method comprising:

accessing a first related object and a second related object

calculating a relationship score for a relationship between the first and second related objects;

selecting a relationship type, using the relationship score, representing the relationship between the first and second related objects, the selected relationship type including at least one of a complement relationship type, indicating complementary relationship between the first and second related objects, and a substitute relation type, indicating a substitutionary relationship between the first and second related objects;

storing the selected relationship type representing the relationship between the first related object and the second related object; and

using the selected relationship type to provide a recommendation within the network-based system.

2. The method of claim 1, wherein the calculating of the relationship score includes calculating a relevance score representing an historic transactional relationship between the first related object and the second related object using historical transaction data associated with the first related object and the second related object.

3. The method of claim 1, wherein the calculating of the relationship score includes:

accessing navigation history data from the network-based system; and

calculating a navigation linkage score representing an amount of navigation between the first related object and the second related object using the navigation history data.

4. The method of claim 1, wherein the calculating of the relationship score includes:

calculating a relevance score representing a transactional history relationship between the first related object and the second related object using historical transaction data associated with the first related object and the second related object;

accessing navigation history data from the network-based system; and

calculating a navigation linkage score representing an amount of navigation between the first related object and the second related object using the navigation history data.

5. The method of claim 4, wherein the selecting of the relationship type includes:

calculating a category ratio using the relevance score and the navigation linkage score;

dynamically computing a complement threshold using the historical data associated with the first related object and the second related object;

dynamically computing a substitute threshold using the historical data associated with the first related object and the second related object; and

selecting the complement relationship type, based on the category ratio transgressing the complement threshold; and

selecting the substitute relation type based on the category ratio transgressing the substitute threshold.

6. The method of claim 1, wherein the first related object and the second related object are one of the following types of related objects:

an item listing;

or

a category of item listings.

7. A method of creating relationships within a network-based system for making recommendations, the method comprising:

accessing a first related object and a second related object;

accessing historical transaction data generated by the network-based system associated with the first related object and the second related object;

calculating a relevance score representing a purchase history relationship between the first and second related objects using the historical transaction data associated with the first and second related objects;

accessing navigation history data generated by the network-based system associated with the first related object and the second related object;

calculating a navigation linkage score representing an amount of navigation between the first and second related objects using the navigation history data;

calculating a relationship score for a relationship between the first and second related objects using the relevance score and the navigation linkage score; and

selecting a relationship type representing the relationship between the first and second related objects using the relationship score;

storing the selected relationship type representing the relationship between the first related object and the second related object; and

providing a recommendation within the network-based system using the stored relationship type.

8. The method of claim 7, wherein the selecting of the relationship type includes selecting a relationship type from relationship types including:

complements indicating a complementary relationship between the first related object and the second related object; and

substitutes indicating a substitutionary relationship between the first related object and the second related object.

9. The method of claim 7, wherein the selecting of the first and second related objects includes selecting from one of the following types of related objects:

a product listing;

a service listing;

or

a category of listings;

10. The method of claim 7, further comprising:

tracking user activity within the network-based system;

detecting a recommendation opportunity based on the user activity tracked within the network-based system, the recommendation opportunity being associated with the first related object;

determining a purchase lifecycle location based on the user activity and the recommendation opportunity;

determining a recommendation relationship type based on the purchase lifecycle location;

selecting a recommendation based on the recommendation relationship type and the relationship score; and

presenting the recommendation.

11. A computer-implemented recommendation system for use within a network-based system, the system comprising:

a relationship module to calculate a relationship score for a relationship between a first related object and a second related object within the network-based system;

a type module to select a relationship type, using the relationship score calculated by the relationship module, representing the relationship between the first related object and the second related object, the selected relationship type including at least one of a complement relationship type, indicating a complementary relationship between the first and second related objects, and a substitute relationship type, indicating a substitutionary relationship between the first and second related objects

a recommendation engine to provide recommendations within the network-based system using the relationship type.

12. The computer-implemented recommendation system of claim 11, wherein the relationship module is to calculate the relationship score by:

calculating a relevance score representing a transactional history relationship between the first related object and the second related object using historical transaction data associated with the first related object and the second related object;

accessing navigation history data from the network-based system; and

calculating a navigation linkage score representing an amount of navigation between the first related object and the second related object using the navigation history data.

13. The computer-implemented recommendation system of claim 12, wherein the type module is to select the relationship type by:

calculating a category ratio using the relevance score and the navigation linkage score;

computing a complement threshold dynamically using the historical data associated with the first related object and the second related object;

computing a substitute threshold dynamically using the historical data associated with the first related object and the second related object; and

selecting the complement relationship type, based on the category ratio transgressing the complement threshold; and

selecting the substitute relation type based on the category ratio transgressing the substitute threshold.

14. The computer-implemented recommendation system of claim 11, further comprising an activity tracking module to track user activity within the network-based system;

and

wherein the recommendation engine is further to;

detect a recommendation opportunity based on the user activity tracked within the network-based system, the recommendation opportunity associated with the first related object;

determine a purchase lifecycle location based on the user activity and the recommendation opportunity; and

select a recommendation based on the recommendation relationship type and the relationship score.

15. The computer-implemented recommendation system of claim 14, wherein the first related object and the second related object are one of the following types of related objects:

an item listing;

or

a category of item listings.

16. A machine-readable storage medium embodying instructions which, when executed by a computer-implemented network-based system, cause the computer-implemented network-based system to execute a method comprising:

accessing a first related object and a second related object

calculating a relationship score for a relationship between the first and second related objects;

selecting a relationship type, using the relationship score, representing the relationship between the first and second related objects, the selected relationship type including at least one of a complement relationship type, indicating complementary relationship between the first and second related objects, and a substitute relation type, indicating a substitutionary relationship between the first and second related objects;

storing the selected relationship type representing the relationship between the first related object and the second related object; and

using the selected relationship type to provide a recommendation within the network-based system.

17. The computer-readable storage medium of claim 16, wherein the calculating of the relationship score includes calculating a relevance score representing an historic transactional relationship between the first related object and the second related object using historical transaction data associated with the first related object and the second related object.

18. The computer-readable storage medium of claim 16, wherein the calculating of the relationship score includes:

accessing navigation history data from the network-based system; and

calculating a navigation linkage score representing an amount of navigation between the first related object and the second related object using the navigation history data.

19. The computer-readable storage medium of claim 16, wherein the calculating of the relationship score includes:

calculating a relevance score representing a transactional history relationship between the first related object and the second related object using historical transaction data associated with the first related object and the second related object;

accessing navigation history data from the network-based system; and

calculating a navigation linkage score representing an amount of navigation between the first related object and the second related object using the navigation history data.

20. The computer-readable storage medium of claim 19, wherein the selecting of the relationship type includes:

calculating a category ratio using the relevance score and the navigation linkage score;

computing a complement threshold dynamically using the historical data associated with the first related object and the second related object;

computing a substitute threshold dynamically using the historical data associated with the first related object and the second related object; and

selecting the complement relationship type, based on the category ratio transgressing the complement threshold; and

selecting the substitute relation type based on the category ratio transgressing the substitute threshold.

21. The computer-readable storage medium of claim 20, further comprising:

tracking user activity within the network-based system;

detecting a recommendation opportunity based on the user activity tracked within the network-based system, the recommendation opportunity being associated with the first related object;

determining a purchase lifecycle location based on the user activity and the recommendation opportunity;

determining a recommendation relationship type based on the purchase lifecycle location;

selecting a recommendation based on the recommendation relationship type and the relationship score; and

presenting the recommendation.

22. The computer-readable storage medium of claim 16, wherein the first related object and the second related object are one of the following types of related objects:

an item listing;

or

a category of item listings.