FLEXIBLE ASSET AND SEARCH RECOMMENDATION ENGINES

• US 20080270250A1
• Filed: 04/26/2007
• Published: 10/30/2008
• Est. Priority Date: 04/26/2007
• Status: Active Grant

• Current Assignee: eBay Inc.
• Original Assignee: eBay Inc.
• Inventors: Mohan, Sunil, Bolivar, Alvaro
• Granted Patent: US 8,050,998 B2
• Time in Patent Office: Days
• US Class Current: 705/26
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0601 Electronic shopping [e-shopping]
• CPC Class Codes: G06Q 30/0623 Item investigation
• CPC Class Codes: G06Q 40/00 Finance; Insurance; Tax strategies; Processing of corporate or income taxes
• CPC Class Codes: G06Q 40/04 Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange

Abstract

The subject mater herein relates to Internet-based commerce and, more particularly, to a flexible asset and search recommendation engines. Various embodiments herein provide systems, methods, and software to identify relationships between assets and provide asset recommendations as a function of the identified relationships. In some embodiments, relationships between assets are identified by gathering statistics including a probability of an action in view of one or more other actions an entity may have performed.

Claims

1. A method comprising:

collecting transaction data of actions with regard to assets performed by entities;

storing the transaction data in a database;

executing a data-mining process against the transaction data to identify relationships between two or more action-asset pairs performed by entities; and

storing a representation of each relationship identified by the data-mining process in an asset recommendation database.

2. The method of claim 1, wherein one or more entities is a registered user of a user account.

3. The method of claim 1, wherein one or more of the actions includes a purchase of an asset.

4. The method of claim 3, wherein identifying the relationship between action-asset pairs includes identifying a relationship between the purchases of the asset as a function of one or more asset specific attributes.

5. The method of claim 3, wherein identifying the relationship between action-asset pairs includes identifying a relationship between the purchase of the assets as a function of a seller of the asset.

6. The method of claim 1, wherein one or more assets includes a product offered for sale.

7. The method of claim 6, wherein the action is a view of an a document categorized as an asset review, guide, and other documentation.

8. The method of claim 6, wherein the action is receiving input indicating an entity owns a particular asset.

9. The method of claim 6, wherein the action is an addition of the asset to a shopping list.

10. The method of claim 6, wherein the action is an addition of the asset to a wish list

11. The method of claim 1, wherein a relationship is identified between an entity action of adding an asset associated with a seller to a favorites list and the identified asset is one or more other products offered by the same seller.

12. The method of claim 1, wherein an asset includes one or more keywords and the data-mining process identifies relationships between action-asset pairs as a function of relationships between the one or more keywords and entity actions involving one or more of the keywords.

13. The method of claim 12, wherein at least one of the entity actions involving a keyword is the representation of a search including one or more keywords stored in an entity search history.

14. The method of claim 12, wherein at least one of the entity actions involving a keyword is the addition of an asset associated to at least one keyword to a shopping list.

15. The method of claim 12, wherein at least one of the entity actions involving a keyword is the addition of an asset associated to at least one keyword to a wish list.

16. The method of claim 1, wherein an asset is associated with one or more keywords and the data-mining process identifies relationships between action-asset pairs as a function of associations of assets to categories and entity actions against assets assigned to the categories.

17. The method of claim 16, wherein the entity action is an addition of an asset associated with a category to a shopping list.

18. The method of claim 16, wherein the entity action is an addition of an asset associated with a category to a wish list.

19. The method of claim 1, wherein the database is a data warehouse.

20. The method of claim 1, wherein a relationship between action-asset pairs performed by the entities is identified as a probability of a combination of the two or more action asset pairs being performed by an entity that performed at least one of the two or more action-asset pairs.

21. The method of claim 1, further comprising:

executing an asset recommendation process, wherein the asset recommendation process;

receives an asset recommendation request including one or more retrieval arguments;

retrieves one or more asset representations from the asset recommendation database as a function of the one or more retrieval arguments; and

returns the retrieved asset representations.

22. The method of claim 1, further comprising:

collecting entity demographic data and storing the demographic data in the database; and

wherein the data mining process further executes against the demographic data and transaction data to identify relationships between one or more entity demographic characteristics represented in the demographic data and one or more action-asset pairs.

25. The method of claim 9, wherein the one or more asset records includes a product offered for sale.

26. The method of claim 25, wherein at least one product offered for sale is a service.

27. The method of claim 25, wherein the data-mining process identifies assets for which an entity has given negative feedback and identifies a comparable product which has received positive feedback.

28. The method of claim 27, further comprising:

executing an asset recommendation process, wherein the asset recommendation process;

receives an asset recommendation request including an entity identifier received as a retrieval argument;

retrieves one or more asset representations from the asset recommendation database as a function of one or more retrieval arguments; and

returns the retrieved asset representations.

23. A method comprising:

maintaining a database including;

asset records, each asset record defining an asset and data descriptive of one or more asset attributes;

entity records, each entity record including data descriptive of one or more entity attributes; and

feedback records, each feedback record including a relationship of an entity record to an asset record and feedback data given by the entity represented by the entity record about the asset of the asset record; and

executing a data-mining process against the feedback records to identify relationships of two or more entity-asset pairs between feedback records; and

storing a representation of the identified relationships in an asset recommendation database.

24. The method of claim 23, wherein:

a feedback record includes an indication of a degree of entity satisfaction with at least one identified attribute of the asset of the feedback record; and

the data-mining process identifies relationships between the two or more entity-asset pairs as a function of the degree of entity satisfaction with the asset of the feedback records.

29. A method comprising:

maintaining a database including;

asset records, each asset record including data descriptive of one or more asset attributes;

entity records, each entity record including data descriptive of one or more entity attributes;

transaction records, each transaction record including a relationship to one or more entity records, a relationship to an asset record, data of one or more actions with regard to the asset record, each action associated to one of the one or more entity records; and

feedback records, each feedback record including;

a relationship to a transaction record; and

feedback data provided by one of the entities represented by an entity record associated with the transaction record, the feedback data including a degree of the entity'"'"'s satisfaction with the asset of the transaction record; and

executing an asset satisfaction data-mining process against the feedback records to determine an overall level of entity satisfaction with at least some of the assets for which feedback has been received;

executing an asset-relation data-mining process against the transaction records to identify relationships between two or more action-asset pairs performed by entities, wherein a relationship between action-asset pairs performed by the entities is identified as a probability of a combination of the two or more action-asset pairs being performed by an entity that performed at least one of the two or more action-asset pairs; and

weighting each probability of the combination of the two or more action-asset pairs by a determined degree of satisfaction with the respective asset as determined by the asset satisfaction data-mining process.

30. The method of claim 29, wherein one or more entities is a registered user of a user account.

31. The method of claim 29, wherein one or more of the actions includes a purchase of an asset.

32. The method of claim 29, wherein one or more assets includes a product offered for sale.

33. The method of claim 29, wherein the database is a data warehouse.

34. The method of claim 29, wherein representations of the relationships identified by the asset-relation data-mining process are stored in an asset recommendation database with the weighted probabilities.

35. The method of claim 34, further comprising:

executing an asset recommendation process, wherein the asset recommendation process;

receives an asset recommendation request including one or more action-asset pairs received as retrieval arguments;

retrieves one or more asset representations from the asset recommendation database as a function of the retrieval arguments; and

returns the retrieved asset representations.

36. The method of claim 29, further comprising:

collecting entity demographic data and storing the demographic data in the database; and

wherein the asset-relation data mining process further executes against the demographic data and transaction data to identify relationships between one or more entity demographic characteristics represented in the demographic data and one or more action-asset pairs.

37. The method of claim 29, further comprising:

executing a replacement product recommendation process that includes;

identifying an entity action-asset pair of an asset where the action indicates the entity owns the asset and the entity provided negative feedback regarding the asset;

identifying assets similar to the asset for which the entity provided negative feedback; and

providing a recommendation of the similar assets to the entity.

38. A method comprising:

maintaining a database including;

asset records, each asset record including data descriptive of one or more asset attributes;

entity records, each entity record including data descriptive of one or more entity attributes;

entity watch list records, wherein;

an entity watch list record is associated with an entity record and an asset record; and

a watch list record is added by an entity to cause data and notifications about the associated asset record to be recorded for and sent to the entity; and

transaction records including an association to an asset and one or more entities; and

executing a data-mining process against the watch list records and transaction records to identify relationships between watch list records and one or more other asset records.

39. A method comprising:

maintaining an asset database including asset representations, category representations, and an association of at least some of the product records to one or more category records;

maintaining entity records, each entity record including data descriptive of one or more entity attributes;

collecting transaction data of entity actions with regard to assets;

executing an asset recommendation process operable to;

identify an asset category from which an entity'"'"'s actions indicate an interest;

identify one or more assets within the identified asset category to recommend to the entity; and

providing the one or more asset recommendations to a recommendation requestor.

40. A method comprising:

maintaining a database of searches submitted by entities;

assigning a score to each search as a function of entity actions taken as a result of respective search recommendations;

receiving a search recommendation request including one or more arguments;

retrieving candidate searches as a function of the one or more arguments and ranking the candidate searches by search score; and

providing one or more search recommendations including a candidate search having a highest score of the candidate searches.

41. The method of claim 40, wherein the arguments included with a search recommendation request include a representation of a previous search and an indication to either broaden or narrow the previous search.

42. The method of claim 40, wherein a score is assigned to each search as a function of one or more actions taken as a result of the search by another entity.

43. The method of claim 42, wherein the one or more actions may include one or both of a purchase and a bid.

44. The method of claim 42, wherein one or more action types are weighted the search scores are aggregated.