Sorting media objects by similarity

• US 7,840,568 B2
• Filed: 05/16/2006
• Issued: 11/23/2010
• Est. Priority Date: 05/16/2006
• Status: Active Grant

• Current Assignee: Sony Corporation (Sony Group Corp.), Sony Electronics Inc. (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.), Sony Electronics Inc. (Sony Group Corp.)
• Inventors: Plutowski, Mark, Purang, Khemdut
• Primary Examiner(s): Wong; Don
• Assistant Examiner(s): PHAN, TUANKHANH D
• Application Number:
  US11/435,494
• Publication Number:
  US 20070271296A1
• Time in Patent Office: 1,652 Days
• Field of Search: 707/748, 707/751, 707/752, 707/737, 709/200, 709/206
• US Class Current: 707/748
• CPC Class Codes: G06F 16/435 Filtering based on additional data, e.g. user or group profiles

Abstract

Weights are assigned for attributes of multimedia objects by sorting the attributes into preference levels, and computing a weight for each preference level. A similarity value of a multimedia object to an object of interest is computed based on the attribute weights.

Claims

1. A computerized method comprising:

sorting, with a recommendation computer, attributes of multimedia objects into preference levels according to an ordering of the attributes, the ordering determined by analysis of past user activity;

computing a weight for each preference level;

computing an attribute weight for each of the attributes as a function of a corresponding preference level weight and a maximum possible number of values for a corresponding attribute, wherein the maximum possible number of values is a multiplicative factor in the computing of the attribute weight; and

computing a similarity value between one of the multimedia objects and another multimedia object based on the attribute weights and a number of matches between the values of attributes common to the one of the multimedia objects and the another multimedia object.

2. The computerized method of claim 1, wherein the ordering represents a user'"'"'s preference of multimedia object attributes.

3. The computerized method of claim 1, further comprising:

ranking objects within the multimedia objects according to the computed similarity value of the multimedia objects to the object of interest.

4. The computerized method of claim 3, further comprising:

outputting a sorted list of ranked multimedia objects.

5. The computerized method of claim 1, wherein the similarity value is further computed using the attribute weights of the attributes of the object to be ranked and a total number of values of the attributes of the object to be ranked.

9. The machine-readable storage medium of claim 1, wherein the similarity value is further computed using the attribute weights of the attributes of the object to be ranked and a total number of values of the attributes of the object to be ranked.

6. A machine-readable storage medium having executable instructions to cause a processor to perform a method, the method comprising:

sorting attributes of multimedia objects into preference levels according to an ordering of the attributes, the ordering determined by analysis of past user activity;

computing a weight for each preference level;

computing an attribute weight for each of the attributes as a function of a corresponding preference level weight and a maximum possible number of values for a corresponding attribute, wherein the maximum possible number of values is a multiplicative factor in the computing of the attribute weight; and

computing a similarity value between one of the multimedia objects and another multimedia object based on the attribute weights and a number of matches between the values of attributes common to the one of the multimedia objects and the another multimedia object.

7. The machine-readable storage medium of claim 6, wherein the ordering represents a user'"'"'s preference of multimedia object, attributes.

8. The machine-readable storage medium of claim 6, wherein the method further comprises:

ranking objects within the multimedia objects according to the computed similarity value of the multimedia objects to the object of interest; and

outputting a sorted list of ranked multimedia objects.

10. A computer system comprising:

a processor coupled to a memory through a bus; and

a process executed from the memory by the processor to cause the processor to

sort attributes of multimedia objects into preference levels according to an ordering of the attributes, the ordering determined by analysis of past user activity,

compute a weight for each preference level,

compute an attribute weight for each of the attributes as a function of a corresponding preference level weight and a maximum possible number of values for a corresponding attribute, wherein the maximum possible number of values is a multiplicative factor in the computing of the attribute weight, and

compute a similarity value between one of the multimedia objects and another multimedia object of interest based on the attribute weights and a number of matches between the values of attributes common to the one of the multimedia objects and the another multimedia object.

11. The computer system of claim 10, wherein the ordering represents a user'"'"'s preference of multimedia object attributes.

12. The computer system of claim 10, wherein the process further causes the processor to:

rank objects within the multimedia objects according to the computed similarity value of the multimedia objects to the object of interest; and

output a sorted list of ranked multimedia objects.

13. The computer system of claim 10, wherein the similarity value is further computed using the attribute weights of the attributes of the object to be ranked and a total number of values of the attributes of the object to be ranked.

14. An apparatus comprising:

means for sorting attributes of multimedia objects into preference levels according to an ordering of the attributes, the ordering determined by analysis of past user activity;

means for computing a weight for each preference level;

means for computing an attribute weight for each of the attributes as a function of a corresponding preference level weight and a maximum possible number of values for a corresponding attribute, wherein the maximum possible number of values is a multiplicative factor in the computing of the attribute weight; and

means for computing a similarity value between one of the multimedia objects and another multimedia object of interest based on the attribute weights and a number of matches between the values of attributes common to the one of the multimedia objects and the another multimedia object.

15. The apparatus of claim 14, further comprising:

means for ranking objects within the multimedia objects according to the computed similarity value of the multimedia objects to the object of interest; and

means for outputting a sorted list of ranked multimedia objects.

16. The apparatus of claim 14, wherein the ordering represents a user'"'"'s preference of multimedia object attributes.