Collaborative filtering engine

• US 8,694,514 B2
• Filed: 10/06/2010
• Issued: 04/08/2014
• Est. Priority Date: 10/12/2009
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Frieden, Kurt, Rudominer, Mitch
• Primary Examiner(s): Alam, Hosain
• Assistant Examiner(s): JOHNSON, JOHNESE T
• Application Number:
  US12/899,406
• Publication Number:
  US 20110087678A1
• Time in Patent Office: 1,280 Days
• Field of Search: 707/749, 707/754
• US Class Current: 707/749
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/954 Navigation, e.g. using categorised browsing

Abstract

Various systems and methods for collaborative filtering of enterprise objects are described. A first object and a second object, which may be enterprise objects, may be compared. Vectors associated with each object may be created. These vectors may contain elements that indicate one or more actions performed involving other objects. Using these vectors, a similarity comparison may be performed to determine the amount of similarity between the objects. Based on a similarity value determined from the similarity comparison between the vectors, the second object may be recommended to a user.

Claims

1. A method for collaborative filtering of enterprise objects, the method comprising:

identifying, by a computer system, a first object;

identifying, by the computer system, a second object to be compared to the first object for similarity, wherein;

the first object and the second object are enterprise objects;

the first and second objects are either users or items; and

the enterprise is a business organization;

creating, by the computer system, a first vector, comprising a first plurality of elements, linked to the first object, wherein each of the first plurality of elements represents at least one user action performed involving the first object by a first set of one or more users;

creating, by the computer system, a second vector, comprising a second plurality of elements, linked to the second object, wherein each of the second plurality of elements represents at least one user action performed involving the second object by a second set of one or more users;

performing, by the computer system, a similarity comparison using the first vector and the second vector to calculate a similarity value that identifies an amount of similarity between user actions involving the first object and user actions involving the second object, wherein the similarity comparison is based at least in part on one or more manners of user interaction corresponding to one or more of;

accessing, handling, qualifying, and/or associating with additional information; and

presenting, by the computer system, an indication of the second object to a user, wherein the indication is at least partially based on the similarity value between the first object and the second object.

2. The method of claim 1, wherein the first object is a first user and the second object is a second user.

3. The method of claim 1, wherein the first object is a first item and the second object is a second item.

4. The method of claim 1, wherein the user has interacted with the first object and the second object is similar to the first object.

5. The method of claim 2, further comprising presenting, by the computer system, an indication of an item to the user, wherein the item has been interacted with by the second user.

6. The method of claim 1, further comprising storing, by the computer system, the similarity value in a database that stores a plurality of similarity values for a plurality of pairs of objects of the enterprise.

7. A method for collaborative filtering of enterprise objects, the method comprising:

identifying, by a computer system, a first object;

identifying, by the computer system, a second object to be compared to the first object, wherein the first object and the second object are part of an enterprise;

creating, by the computer system, a first vector linked to the first object, wherein the first vector comprises a first plurality of elements representing user actions performed involving the first object by a first set of one or more users;

creating, by the computer system, a second vector linked to the second object, wherein the second vector comprises a second plurality of elements representing user actions performed involving the second object by a second set of one or more users;

performing, by the computer system, a similarity comparison using the first vector and the second vector, wherein the similarity comparison calculates a similarity value that identifies an amount of similarity between the first object and the second object of the enterprise based on user actions performed involving the first object and the second object, wherein the similarity comparison is based at least in part on one or more manners of user interaction corresponding to one or more of;

accessing, handling, qualifying, and/or associating with additional information.

8. The method of claim 7, wherein the first object is a first user and the second object is a second user.

9. The method of claim 7, wherein the first object is a first item and the second object is a second item.

10. The method of claim 7, wherein the similarity comparison comprises using a Tanimoto similarity function and the first vector and second vector are binary vectors.

11. The method of claim 10, wherein the Tanimoto similarity function is a generalized Tanimoto similarity function and the first vector and second vector are non-negative vectors.

12. The method of claim 9, wherein:

the first vector comprises a first element of the first plurality of elements that identifies whether a first user has liked the first item; and

the second vector comprises a second element of the second plurality of elements that identifies whether the first user has liked the second item.

13. The method of claim 7, further comprising presenting, by the computer system, a recommendation to a user, wherein:

the recommendation identifies the second object; and

the recommendation is at least partially based on the similarity value that identifies the amount of similarity between the first object and the second object of the enterprise.

14. The method of claim 7, further comprising storing, by the computer system, the similarity value that identifies the amount of similarity in a database that stores a plurality of similarity values for a plurality of pairs of objects of the enterprise.

15. The method of claim 7, wherein performing, by the computer system, the similarity comparison using the first vector and the second vector further comprises:

determining, by the computer system, a first number based on elements of the first vector and elements of the second vector that indicate at least one same action has been performed using the first object and the second object;

determining, by the computer system, a second number based on elements in the first vector and elements of the second vector, wherein at least one element of the first vector and the second vector indicates an action has been performed using at least one of the first object and the second object; and

calculating, by the computer system, using the first number and the second number, the similarity value that identifies the amount of similarity between the first object and the second object of the enterprise based on actions involving the first object and the second object.

16. A computer program product stored on a non-transitory computer-readable storage medium for collaborative filtering of enterprise objects, the computer program product comprising instructions for:

identifying a first object;

identifying a second object to be compared with the first object, wherein the first object and the second object are part of an enterprise;

creating a first vector linked to the first object, wherein the first vector comprises a first plurality of elements representing user actions performed involving the first object by a first set of one or more users;

creating a second vector linked to the second object, wherein the second vector comprises a second plurality of element representing user actions performed involving the second object by a second set of one or more users; and

performing a similarity comparison using the first vector and the second vector, wherein the similarity comparison calculates a similarity value that identifies an amount of similarity between the first object and the second object of the enterprise based on user actions performed involving the first object and the second object, wherein the similarity comparison is based at least in part on one or more manners of user interaction corresponding to one or more of;

accessing, handling, qualifying, and/or associating with additional information.

17. The computer program product of claim 16, wherein the first object is a first user and the second object is a second user.

18. The computer program product of claim 16, wherein the first object is a first item and the second object is a second item.

19. The computer program product of claim 16, wherein the similarity comparison comprises using a Tanimoto similarity function.

20. The computer program product of claim 17, wherein:

the first vector comprises a first element of the first plurality of elements that identifies whether the first user has performed an action involving a first item; and

the second vector comprises a second element of the second plurality of elements that identifies whether the second user has performed the action involving the first item.