SYSTEMS AND METHODS FOR RECOGNIZING AMBIGUITY IN METADATA

• US 20130332400A1
• Filed: 06/07/2013
• Published: 12/12/2013
• Est. Priority Date: 06/08/2012
• Status: Active Grant

• Current Assignee: Spotify AB (Spotify Technology SA)
• Original Assignee: Spotify AB (Spotify Technology SA)
• Inventors: Gonzlez, Alejandro Machado
• Granted Patent: US 9,230,218 B2
• Time in Patent Office: Days
• US Class Current: 706/12
• CPC Class Codes: G06F 16/48 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 16/686 using information manually generated, e.g. tags, keywords, comments, title or artist information, time, location or usage information, user ratings
• CPC Class Codes: G06N 5/048 Fuzzy inferencing
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks

Abstract

A method for estimating artist ambiguity in a dataset is performed at a device with a processor and memory storing instructions for execution by the processor. The method includes applying a statistical classifier to a first dataset including a plurality of media items, wherein each media item is associated with one of a plurality of artist identifiers, each artist identifier identifies a real world artist, and the statistical classifier calculates a respective probability that each respective artist identifier is associated with media items from two or more different real world artists based on a respective feature vector corresponding to the respective artist identifier. The method further includes providing a report of the first dataset, including the calculated probabilities, to a user of the electronic device. Each respective feature vector includes a plurality of features that indicate likelihood of artist ambiguity.

Claims

1. A method for estimating artist ambiguity in a dataset, comprising:

at an electronic device having one or more processors and memory storing one or more programs for execution by the one or more processors;

applying a statistical classifier to a first dataset including a plurality of media items, wherein each media item is associated with one of a plurality of artist identifiers, each artist identifier identifies a real world artist, and the statistical classifier calculates a respective probability that each respective artist identifier is associated with media items from two or more different real world artists based on a respective feature vector corresponding to the respective artist identifier; and

providing a report of the first dataset, including the calculated probabilities, to a user of the electronic device;

wherein each respective feature vector includes features selected from the group consisting of;

whether the corresponding respective artist identifier matches multiple artist entries in one or more second datasets;

whether a respective number of countries of registration of media items associated with the corresponding respective artist identifier exceeds a predetermined country threshold;

whether a respective number of characters in the corresponding respective artist identifier exceeds a predetermined character threshold;

whether a respective number of record labels associated with the corresponding respective artist identifier exceeds a predetermined label threshold;

whether the corresponding respective artist identifier is associated with albums in at least two different languages; and

whether a difference between an earliest release date and a latest release date of media items associated with the corresponding respective artist identifier exceeds a predetermined time span threshold.

2. The method of claim 1, wherein the statistical classifier is a naï

ve Bayes classifier.

3. The method of claim 2, wherein the naï

ve Bayes classifier takes the form

4. The method of claim 3, wherein the feature vector includes the following features:

whether the corresponding respective artist identifier matches multiple artist entries in one or more second datasets;

whether a respective number of characters in the corresponding respective artist identifier exceeds a predetermined character threshold; and

whether the corresponding respective artist identifier is associated with albums in at least two different languages.

5. The method of claim 1, wherein the statistical classifier is a logistic regression classifier.

6. The method of claim 4, wherein the logistic regression classifier takes the form

7. The method of claim 6, wherein the feature vector includes the following features:

whether the corresponding respective artist identifier matches multiple artist entries in one or more second datasets;

whether a respective number of countries of registration of media items associated with the corresponding respective artist identifier exceeds a predetermined country threshold;

whether a respective number of characters in the corresponding respective artist identifier exceeds a predetermined character threshold; and

whether the corresponding respective artist identifier is associated with albums in at least two different languages.

8. The method of claim 1, further comprising:

determining whether each respective probability satisfies a predetermined probability condition; and

setting a flag for each respective artist identifier that satisfies the predetermined probability condition.

9. The method of claim 8, wherein the predetermined probability condition is whether the respective probability exceeds a predetermined probability threshold.

10. The method of claim 9, wherein the probability threshold is 0.5.

11. The method of claim 9, wherein the probability threshold is 0.9.

12. The method of claim 1, wherein the predetermined country threshold is 1 country.

13. The method of claim 1, wherein the predetermined character threshold is 13 characters.

14. The method of claim 1, wherein the predetermined label threshold is 3 record labels.

15. A method for estimating artist ambiguity in a dataset, comprising:

at an electronic device having one or more processors and memory storing one or more programs for execution by the one or more processors;

applying a statistical classifier to a first dataset including a plurality of media items, wherein each media item is associated with one of a plurality of artist identifiers, each artist identifier identifies a real world artist, and the statistical classifier calculates a respective probability that each respective artist identifier is associated with media items from two or more different real world artists based on a respective feature vector corresponding to the respective artist identifier; and

providing a report of the first dataset, including the calculated probabilities, to a user of the electronic device;

wherein each respective feature vector includes features selected from the group consisting of;

whether the corresponding respective artist identifier matches multiple artist entries in one or more second datasets;

a respective number of countries of registration of media items associated with the corresponding respective artist identifier;

a respective number of characters in the corresponding respective artist identifier;

a respective number of record labels associated with the corresponding respective artist identifier;

a respective number of languages of albums associated with the corresponding respective artist identifier; and

a respective difference between an earliest release date and a latest release date of media items associated with the corresponding respective artist identifier.

16. A method for estimating artist ambiguity in a dataset, comprising:

at an electronic device having one or more processors and memory storing one or more programs for execution by the one or more processors;

applying a statistical classifier to a first dataset including a plurality of media items, wherein each media item is associated with one of a plurality of artist identifiers, each artist identifier identifies a real world artist, and the statistical classifier calculates a respective probability that each respective artist identifier is associated with media items from two or more different real world artists based on a respective feature vector corresponding to the respective artist identifier.

17. The method of claim 16, wherein each respective feature vector includes features selected from the group consisting of:

whether the corresponding respective artist identifier matches multiple artist entries in one or more second datasets;

whether a respective number of countries of registration of media items associated with the corresponding respective artist identifier exceeds a predetermined country threshold;

whether a respective number of characters in the corresponding respective artist identifier exceeds a predetermined character threshold;

whether a respective number of record labels associated with the corresponding respective artist identifier exceeds a predetermined label threshold;

whether the corresponding respective artist identifier is associated with albums in at least two different languages; and

whether a difference between an earliest release date and a latest release date of media items associated with the corresponding respective artist identifier exceeds a predetermined time span threshold.

18. The method of claim 17, further comprising providing a report of the first dataset, including the calculated probabilities, to a user of the electronic device.

19. The method of claim 17, further comprising:

determining whether the respective probabilities of the respective artist identifiers satisfy a predetermined probability condition; and

in response to detecting that a particular probability of a particular artist identifier satisfies the predetermined probability condition;

creating a new artist identifier; and

associating one or more particular media items with the new artist identifier, wherein the one or more particular media items were previously associated with the particular artist identifier.

20. The method of claim 19, further comprising, prior to creating the new artist identifier, identifying the one or more media items in a second dataset by identifying a first artist entry in the second dataset that is associated with the one or more media items and has a same artist name as the particular artist identifier, and identifying a second artist entry in the second dataset that is not associated with the one or more media items and has the same name as the particular artist identifier.

21. A computer system, comprising:

one or more processors; and

memory storing one or more programs for execution by the one or more processors, the one or more programs including instructions for;

applying a statistical classifier to a first dataset including a plurality of media items, wherein each media item is associated with one of a plurality of artist identifiers, each artist identifier identifies a real world artist, and the statistical classifier calculates a respective probability that each respective artist identifier is associated with media items from two or more different real world artists based on a respective feature vector corresponding to the respective artist identifier; and

providing a report of the first dataset, including the calculated probabilities, to a user of the electronic device;

wherein each respective feature vector includes features selected from the group consisting of;

whether the corresponding respective artist identifier matches multiple artist entries in one or more second datasets;

whether a respective number of countries of registration of media items associated with the corresponding respective artist identifier exceeds a predetermined country threshold;

whether a respective number of characters in the corresponding respective artist identifier exceeds a predetermined character threshold;

whether a respective number of record labels associated with the corresponding respective artist identifier exceeds a predetermined label threshold;

whether the corresponding respective artist identifier is associated with albums in at least two different languages; and

whether a difference between an earliest release date and a latest release date of media items associated with the corresponding respective artist identifier exceeds a predetermined time span threshold.

22. A non-transitory computer readable storage medium storing one or more programs, the one or more programs comprising instructions, which, when executed by a portable electronic device or a computer system with one or more processors, cause the device to:

apply a statistical classifier to a first dataset including a plurality of media items, wherein each media item is associated with one of a plurality of artist identifiers, each artist identifier identifies a real world artist, and the statistical classifier calculates a respective probability that each respective artist identifier is associated with media items from two or more different real world artists based on a respective feature vector corresponding to the respective artist identifier; and

provide a report of the first dataset, including the calculated probabilities, to a user of the electronic device;

wherein each respective feature vector includes features selected from the group consisting of;

whether the corresponding respective artist identifier matches multiple artist entries in one or more second datasets;

whether a respective number of countries of registration of media items associated with the corresponding respective artist identifier exceeds a predetermined country threshold;

whether a respective number of characters in the corresponding respective artist identifier exceeds a predetermined character threshold;

whether a respective number of record labels associated with the corresponding respective artist identifier exceeds a predetermined label threshold;

whether the corresponding respective artist identifier is associated with albums in at least two different languages; and

whether a difference between an earliest release date and a latest release date of media items associated with the corresponding respective artist identifier exceeds a predetermined time span threshold.