Matching media file metadata to standardized metadata

• US 20050055372A1
• Filed: 09/04/2003
• Published: 03/10/2005
• Est. Priority Date: 09/04/2003
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Springer, Thomas Brian Jr., McCartney, Jason E.D.
• Granted Patent: US 7,546,288 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/313 Selection or weighting of terms for indexing
• CPC Class Codes: G06F 16/907 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: Y10S 707/99931 Database or file accessing
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access
• CPC Class Codes: Y10S 707/99942 Manipulating data structure, e.g. compression, compaction, compilation
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing

Abstract

Providing standardized metadata associated with media content responsive to input metadata. The invention extracts one or more tokens from the input metadata. Each of the extracted tokens represents a portion of the input metadata. The invention creates a token group comprising a plurality of selected tokens. The invention searches the database of standardized metadata using a multi-pass approach using the token group and the extracted tokens to identify the standardized metadata associated with the input metadata.

Claims

1. A method for providing metadata associated with media content and stored in a database, said method comprising:

receiving, from a computing device, a request comprising an item of initial metadata identified by a media file, said media file storing media content;

extracting one or more tokens from the item of initial metadata in the received request, each of said extracted tokens representing a portion of the item of initial metadata;

selecting a plurality of the extracted tokens to create a token group; and

searching the database for additional metadata associated with the media content using the created token group.

2. The method of claim 1, wherein searching the database comprises searching the database for additional metadata, said additional metadata representing standardized metadata.

3. The method of claim 1, wherein selecting the plurality of the extracted tokens comprises creating a permutation of the extracted tokens.

4. The method of claim 1, wherein extracting one or more tokens comprises extracting one or more metatags from the item of initial metadata in the received request, each of said metatags specifying textual information.

5. The method of claim 1, wherein the created token group comprises a text string, and wherein searching the database comprises searching the database using a substring of the created token group.

6. The method of claim 1, wherein searching the database produces additional metadata relating to a single item of media content, and further comprising identifying the additional metadata and the single item of media content to the computing device.

7. The method of claim 1, wherein searching the database produces an unranked list of items of media content.

8. The method of claim 7, further comprising sending the unranked list of items of media content to the computing device.

9. The method of claim 7, further comprising:

searching the database using the extracted tokens to produce another list of items of media content;

ranking the other list of items of media content; and

sending the ranked list of items of media content to the computing device.

10. The method of claim 9, wherein ranking the other list of items of media content comprises considering an edit distance between the extracted tokens.

11. The method of claim 9, wherein ranking the other list of items of media content comprises considering the popularity of each of the items of media content in the other list.

12. The method of claim 7, further comprising searching the database using a variation of one or more of the extracted tokens.

13. The method of claim 1, further comprising cleansing the extracted tokens by removing punctuation from the extracted tokens.

14. The method of claim 1, wherein receiving the request comprises receiving the request with a plurality of items of initial metadata, each of said items of initial metadata corresponding to one or more of the following:

album title, artist, track title, and track number.

15. The method of claim 1, wherein receiving the request comprises receiving the request with a plurality of items of initial metadata each corresponding to one of a plurality of categories, and wherein searching the database comprises searching the database based on the items corresponding to one of the categories.

16. The method of claim 15, further comprising performing additional searching based on the items corresponding to at least one of the other categories.

17. The method of claim 15, wherein the categories comprise at least one of the following:

album title, artist name, and song title.

18. The method of claim 1, further comprising assigning a weight to each of the extracted tokens, said assigned weight corresponding to a frequency of occurrence of the extracted token in standardized metadata stored in the database.

19. The method of claim 18, wherein searching the database comprises generating a list of items of media content ordered according to the assigned weights.

20. One or more computer-readable media having computer-executable instructions for performing the method recited in claim 1.

21. A method for providing metadata associated with media content and stored in a database, said method comprising:

receiving, from a computing device, a request comprising metadata associated with media content, said metadata including text strings corresponding to an album title, an artist name, and a track title;

extracting one or more tokens from the text strings, each of said extracted tokens representing a substring of the text strings;

selecting a plurality of the extracted tokens from the album title text string and the artist name text string to create a first token group; and

searching the database for additional metadata associated with the media content using the first token group.

22. The method of claim 21, wherein searching the database produces the additional metadata relating to a single item of media content, and further comprising identifying the additional metadata and the single item of media content to the computing device.

23. The method of claim 21, wherein searching the database yields a plurality of results, and further comprising:

selecting a plurality of the extracted tokens from the album title text string and the track title text string to create a second token group; and

searching the database for additional metadata associated with the media content using the second token group.

24. The method of claim 23, wherein searching the database using the second token group produces the additional metadata relating to a single item of media content, and further comprising identifying the additional metadata and the single item of media content to the computing device.

25. The method of claim 23, wherein searching the database using the second token group yields a plurality of results, and further comprising:

selecting a plurality of the extracted tokens from the artist name text string and the track title text string to create a third token group; and

searching the database for additional metadata associated with the media content using the third token group.

26. The method of claim 23, wherein searching the database using the third token group produces the additional metadata relating to a single item of media content, and further comprising identifying the additional metadata and the single item of media content to the computing device.

27. One or more computer-readable media having computer-executable instructions for performing the method recited in claim 21.

28. One or more computer-readable media having computer-executable components for providing metadata associated with media content and stored in a database, said components comprising:

an interface component for receiving, from a computing device, a request comprising an item of initial metadata identified by a media file, said media file storing media content;

a tokenizer component for extracting one or more tokens from the item of initial metadata in the request received by the interface component, each of said extracted tokens representing a portion of the item of initial metadata, said tokenizer component further selecting a plurality of the extracted tokens to create a token group; and

a query component for searching the database for additional metadata associated with the media content using the token group created by the tokenizer component.

29. The computer-readable media of claim 28, wherein the tokenizer component creates the token group as a text string, and wherein the query component searches the database using a substring of the created text string.

30. The computer-readable media of claim 28, wherein the query component produces additional metadata relating to a single item of media content and further identifies the additional metadata and the single item of media content to the computing device.

31. The computer-readable media of claim 28, wherein the interface component receives the request with a plurality of items of initial metadata, each of said plurality of items of initial metadata corresponding to one or more of the following:

album title, artist, track title, and track number.

32. The computer-readable media of claim 28, wherein the request comprises a plurality of items of initial metadata each corresponding to one of a plurality of categories, and wherein the query component searches the database based on the items corresponding to one of the categories.

33. The computer-readable media of claim 32, wherein the query component further performs additional searching based on the items corresponding to at least one of the other categories.

34. The computer-readable media of claim 32, wherein the categories comprise at least one of the following:

album title, artist name, and song title.

35. A web service receiving a request from a computing device, said request comprising an item of initial metadata identified by a media file, said media file storing media content, said web service comprising:

a tokenizer component for extracting one or more tokens from the item of initial metadata in the received request, each of said extracted tokens representing a portion of the item of initial metadata, said tokenizer component further selecting a plurality of the extracted tokens to create a token group;

a database storing additional metadata relating to the media content; and

a query component for searching the database for the additional metadata associated with the media content using the created token group.

36. The web service of claim 35, wherein the tokenizer component creates the token group as a text string, and wherein the query component searches the database using a substring of the created text string.

37. The web service of claim 35, wherein the query component produces additional metadata relating to a single item of media content and further identifies the additional metadata and the single item of media content to the computing device.

38. The web service of claim 35, wherein the web service receives the request with a plurality of items of initial metadata, each of said plurality of items of initial metadata corresponding to one or more of the following:

album title, artist, track title, and track number.

39. The web service of claim 35, wherein the request comprises a plurality of items of initial metadata each corresponding to one of a plurality of categories, and wherein the query component searches the database based on the items corresponding to one of the categories.

40. The web service of claim 39, wherein the query component further performs additional searching based on the items corresponding to at least one of the other categories.

41. The web service of claim 39, wherein the categories comprise at least one of the following:

album title, artist name, and song title.

42. A computer-readable medium having stored thereon a data structure describing an item of media content, said data structure comprising:

a metadata item associated with the media content;

one or more tokens, each of said tokens representing a portion of the metadata item; and

a token group including one or more of the tokens.

43. The computer-readable medium of claim 42, wherein the token group comprises a permutation of a plurality of the tokens.

44. The computer-readable medium of claim 42, wherein the metadata item comprises one of the following:

an album title, an artist, a track title, and a track number.

45. The computer-readable medium of claim 42, wherein each of the tokens comprises a text string, and further comprising a weight associated with the text string, said weight representing a frequency of occurrence of the text string in the metadata item.

46. A method of processing data for storage on a computer-readable medium, said method comprising:

receiving one or more metadata items from a metadata source, said metadata items being related to media content;

extracting one or more tokens from the received metadata items, each of said extracted tokens representing a portion of one or more of the received metadata items;

selecting a plurality of the extracted tokens to create a token group; and

storing the received metadata items, the extracted tokens, and the created token group on a computer-readable medium to associate the media content with the stored metadata items, tokens, and token group.

47. The method of claim 46, further comprising:

assigning a weight to each of the extracted tokens based on the frequency of occurrence of the extracted token in the metadata items; and

ranking each of the extracted tokens for each of the metadata items based on the assigned weight.

48. The method of claim 46, wherein each of the received metadata items corresponds to a category of metadata, and wherein selecting the plurality of the extracted tokens comprises creating a token group for each category of metadata.

49. The method of claim 46, wherein the metadata items each comprise a text string having one or more words, and wherein extracting one or more tokens comprises extracting each word from the text string.

50. The method of claim 49, wherein extracting one or more tokens further comprises omitting an extracted word based on its level of descriptiveness of the media content.

51. One or more computer-readable media having computer-executable instructions for performing the method recited in claim 46.