Method and system for predicting association item affinities using second order user item associations

• US 9,361,624 B2
• Filed: 03/23/2011
• Issued: 06/07/2016
• Est. Priority Date: 03/23/2011
• Status: Active Grant

• Current Assignee: Ipar LLC
• Original Assignee: Ipar LLC
• Inventors: Spears, Joseph L., Brandstetter, Jeffrey D.
• Primary Examiner(s): DWIVEDI, MAHESH H
• Application Number:
  US13/069,942
• Publication Number:
  US 20120246174A1
• Time in Patent Office: 1,903 Days
• Field of Search: 707/737
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 16/90324 using system suggestions
• CPC Class Codes: G06Q 2220/18 Licensing
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities
• CPC Class Codes: G06Q 30/0246 Traffic
• CPC Class Codes: G06Q 50/01 Social networking

Abstract

Systems and methods are provided for determining an affinity score of a source entity and a target entity without consideration of data that directly links the source entity and the target entity. An intermediate entity that is relatable to a source entity is identified. A target entity is identified that is relatable to an intermediate entity. A similarity score is calculated identifying a degree of similarity between the source entity and the intermediate entity, or the target entity and the intermediate entity. An affinity score is calculated of the source entity and the intermediate entity, or the target entity and the intermediate entity. An affinity score of the source entity and the target entity is determined based on the similarity score and the affinity score.

Claims

1. A computer implemented method of searching for a target entity for which a source entity has greater than a threshold affinity without consideration of data that directly links the source entity and the target entity, comprising:

performing a search for a target entity for which a source entity has greater than a threshold affinity based on a query, wherein performing the search comprises;

identifying, using a computer system, a first intermediate entity and a second intermediate entity that are relatable to a source entity;

identifying, using the computer system, a target entity that is relatable to the first intermediate entity and the second intermediate entity;

determining, using the computer system, a first similarity score identifying a degree of similarity between the source entity and the first intermediate entity, or the target entity and the first intermediate entity;

determining, using the computer system, a second similarity score identifying a degree of similarity between the source entity and the second intermediate entity, or the target entity and the second intermediate entity;

determining a normalization value based on the first similarity score and the second similarity score, wherein the normalization value is determined based on a square root of a sum of squares of the first similarity score and the second similarity score;

calculating, using the computer system, an intermediate affinity score of the source entity and the intermediate entity, or the target entity and the intermediate entity; and

determining, using the computer system, an affinity score of the source entity and the target entity based on the first similarity score, the second similarity score, the intermediate affinity score, and the normalization value; and

displaying an indication of the target entity on a graphical user interface when the affinity score is greater than a threshold level.

2. The method of claim 1, wherein the source entity, the target entity, or the intermediate entity is an association.

3. The method of claim 2, wherein the association is a group to which a plurality of users belong.

4. The method of claim 3, wherein the users have identified themselves as members of the group.

5. The method of claim 3, wherein the users have been identified as members of the group by the system.

6. The method of claim 1, wherein the source entity, the target entity, or the intermediate entity is an item.

7. The method of claim 6, wherein the item is a digital good, a physical good, or a service.

8. The method of claim 7, wherein the digital good includes a video component, an audio component, a text component or a game component.

9. The method of claim 1, wherein the source entity, the target entity, or the intermediate entity is a user.

10. The method of claim 1, wherein an affinity is the actual or predicted preference that a particular user of the system will express toward a particular item.

11. The method of claim 1, wherein an affinity is the actual or predicted preference that the members of a particular association will express toward a particular item.

12. The method of claim 1, wherein the intermediate entity and the source entity are relatable as determined by an affinity between the intermediate entity and the source entity;

wherein the target entity and the intermediate entity are relatable as determined by a similarity between the target entity and the intermediate entity.

13. The method of claim 12, wherein the source entity is an item, the target entity is an association, and the intermediate entity is an association.

14. The method of claim 12, wherein the source entity is an association, the target entity is a first item, and the intermediate entity is a second item.

15. The method of claim 12, wherein the affinity score of the source entity and the target entity is based on an affinity score of the source entity and the intermediate entity or

a similarity score of the intermediate entity and the target entity.

16. The method of claim 15, wherein the similarity score of the intermediate entity and the target entity is determined based on a history of relationships between users and associations.

17. The method of claim 15, wherein the similarity score of the intermediate entity and the target entity is determined based on a history of the interactions between users and items.

18. The method of claim 17, wherein the interactions include an initiation of a purchase transaction by a user.

19. The method of claim 1, wherein the intermediate entity and the source entity are relatable as determined by a similarity between the intermediate entity and the source entity;

wherein the target entity and the intermediate entity are relatable as determined by an affinity between the target entity and the intermediate entity.

20. The method of claim 19, wherein the source entity is an item, the target entity is an association, and the intermediate entity is an item.

21. The method of claim 19, wherein the source entity is an association, the target entity is an item, and the intermediate entity is an association.

22. The method of claim 19, wherein the affinity score of the source entity and target entity is based on an affinity score of the intermediate entity and the target entity and a similarity score of the source entity and the intermediate entity.

23. The method of claim 22, wherein the similarity score of the source entity and intermediate entity is determined based on

selections of the item by users;

or

relationships between the users and an association.

24. The method of claim 22, wherein the affinity score of the intermediate entity and the target entity is determined based on selections of items by users and relationships between users and associations.

25. The method of claim 22, wherein the similarity score of the source entity and intermediate entity is determined based on a history of selections of items by users.

26. The method of claim 1, further comprising identifying a second intermediate entity that is different than the first intermediate entity, wherein the second intermediate entity has a similarity to the first intermediate entity;

and

wherein the target entity is relatable to the second intermediate entity.

27. The method of claim 26, wherein the first and second intermediate entities are users, the target entity is an association, and the target entity and the second intermediate entity are relatable based on the second intermediate entity having membership in the target entity.

28. The method of claim 1, wherein one of the source entity, target entity, or intermediate entity is an item, wherein one of the remaining entities is an association, and wherein an affinity between an item and an association is a probability that users in the association will select the item.

29. The method of claim 1, wherein one of the source entity, target entity, or intermediate entity is an item, wherein one of the remaining entities is a user, and wherein an affinity between an item and a user is a probability that user will select the item.

30. The method of claim 1, wherein an aggregate affinity score of the source entity and the target entity is determined based on additional affinity scores of the source entity and the target entity;

and

wherein the additional affinity scores of the source entity and the target entity are determined using additional intermediate entities.

31. The method of claim 1,

wherein the similarity score is calculated using a term frequency—

inverse data frequency calculation.

32. The method of claim 1, wherein the similarity score is calculated using a cluster table.

33. A computer implemented system for searching for a target entity for which a source entity has greater than a threshold affinity without consideration of data that directly links the source entity and the target entity, comprising:

one or more data processors;

a non-transitory computer-readable medium encoded with instructions for commanding the one or more data processors to execute a method that includes steps of;

performing a search for a target entity for which a source entity has greater than a threshold affinity based on a query, wherein performing the search comprises;

identifying, using a computer system, a first intermediate entity and a second intermediate entity that are relatable to a source entity;

identifying, using the computer system, a target entity that is relatable to the first intermediate entity and the second intermediate entity;

determining, using the computer system, a first similarity score identifying a degree of similarity between the source entity and the first intermediate entity, or the target entity and the first intermediate entity;

determining, using the computer system, a second similarity score identifying a degree of similarity between the source entity and the second intermediate entity, or the target entity and the second intermediate entity;

determining a normalization value based on the first similarity score and the second similarity score, wherein the normalization value is determined based on a square root of a sum of squares of the first similarity score and the second similarity score;

calculating, using the computer system, an intermediate affinity score of the source entity and the intermediate entity, or the target entity and the intermediate entity; and

determining, using the computer system, an affinity score of the source entity and the target entity based on the first similarity score, the second similarity score, the intermediate affinity score, and the normalization value; and

displaying an indication of the target entity on a graphical user interface when the affinity score is greater than a threshold level.

34. A non-transitory computer-readable medium encoded with instructions for commanding one or more data processors to execute a method of searching for a target entity for which a source entity has greater than a threshold affinity without consideration of data that directly links the source entity and the target entity, the method comprising:

performing a search for a target entity for which a source entity has greater than a threshold affinity based on a query, wherein performing the search comprises;

identifying, using a computer system, a first intermediate entity and a second intermediate entity that are relatable to a source entity;

identifying, using the computer system, a target entity that is relatable to the first intermediate entity and the second intermediate entity;

determining, using the computer system, a first similarity score identifying a degree of similarity between the source entity and the first intermediate entity, or the target entity and the first intermediate entity;

determining, using the computer system, a second similarity score identifying a degree of similarity between the source entity and the second intermediate entity, or the target entity and the second intermediate entity;

determining a normalization value based on the first similarity score and the second similarity score, wherein the normalization value is determined based on a square root of a sum of squares of the first similarity score and the second similarity score;

calculating, using the computer system, an intermediate affinity score of the source entity and the intermediate entity, or the target entity and the intermediate entity; and

determining, using the computer system, an affinity score of the source entity and the target entity based on the first similarity score, the second similarity score, the intermediate affinity score, and the normalization value; and

displaying an indication of the target entity on a graphical user interface when the affinity score is greater than a threshold level.