Matching media file metadata to standardized metadata

• US 7,546,288 B2
• Filed: 09/04/2003
• Issued: 06/09/2009
• Est. Priority Date: 09/04/2003
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: McCartney, Jason E. D., Springer, Thomas Brian Jr.
• Primary Examiner(s): Alam; Hosain T
• Assistant Examiner(s): Saeed; Usmaan
• Application Number:
  US10/654,726
• Publication Number:
  US 20050055372A1
• Time in Patent Office: 2,105 Days
• Field of Search: 707/104.1, 707/5, 707/102
• 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 on a first computing device, said method comprising:

receiving, by the first computing device from a second computing device storing a media file storing media content, a request comprising an item of initial metadata identified by the media file, wherein said request is in response to selecting of the media file via a media player by a user of the second computing device;

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;

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;

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

searching the database on the first computing device for additional metadata associated with the media content using the created token group, wherein the token group comprises a text string and searching the database comprises searching the database using a substring of less than the entire text string, said searching comprising generating a list of items of media content ordered according to the assigned weights; and

providing the additional metadata associated with the media content from the database on the first computing device to the second computing device wherein the additional metadata provided to the first computing device comprises the generated list of items of media content.

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 including a plurality of words, and wherein searching the database comprises searching the database using less than all of the plurality of words.

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 second 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 second 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 second 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. One or more computer-readable storage medium having computer-executable instructions for performing the method recited in claim 1.

19. A method for providing standardized metadata associated with media content and stored in a database on a first computing device, said method comprising:

selecting a media file storing media content via a media player, said media file stored on a second computing device, wherein said selecting is performed by a user of the second computing device;

receiving, from the second computing device storing the media file having the media content, a request comprising metadata associated with said media content, said metadata including text strings corresponding to an album title, an artist name, and a track title, wherein said request is in response to selecting of the media file via a media player by a user of the second computing device;

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

assigning a weight to each of the extracted tokens, said assigned weight corresponding to a frequency of occurrence of the external token in the standardized metadata stored in the database;

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;

searching the database on the first computing device for additional metadata associated with the media content using the first token group, said searching comprising generating a list of items of media content ordered according to the assigned weights;

modifying one or more of the extracted tokens to create various keyword groups;

searching the database using the various keyword groups to yield a plurality of results;

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;

searching the database for additional metadata associated with the media content using the second token group to yield another plurality of results;

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;

searching the database for additional metadata associated with the media content using the third token group to produce the additional metadata relating to a single item of media content; and

providing the additional metadata associated with the media content and the single item of media content from the database on the first computing device to the second computing device.

20. One or more computer-readable storage medium having computer-executable instructions for performing the method recited in claim 19.

21. One or more computer-readable storage medium having computer-executable components for providing metadata associated with media content and stored in a database on a first computing device, said components comprising:

a user interface component of a media player for enabling a user of a second computing device to select a media file stored on the second computing device, said media file storing media content;

an interface component for receiving, from the second computing device, a request comprising an item of initial metadata identified by the selected media file stored on said second computing device, said media file storing media content, wherein said request is in response to the user selecting the media file;

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;

a weighting component for assigning a weight to each of the extracted tokens, said assigned weight corresponding to a frequency of occurrence of the extracted token in the metadata stored in the database; and

a query component for searching the database on the first computing device for additional metadata associated with the media content using the token group created by the tokenizer component and for providing the additional metadata from the database on the first computing device to the second computing device, wherein searching the database comprises generating a list of items of media content ordered according to the assigned weights of the extracted tokens, wherein the second computing device updates metadata associated with the media file stored on the second computing device with the additional metadata from the database, 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, wherein the query component further performs additional searching based on the items corresponding to at least one of the other categories, and 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 second computing device.

22. The computer-readable storage medium of claim 21, 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.

23. The computer-readable storage medium of claim 21, 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.

24. The computer-readable storage medium of claim 21, wherein the categories comprise at least one of the following:

album title, artist name, and song title.

25. A system for receiving a request from a first computing device, said request comprising an item of initial metadata identified by a media file stored on said first computing device, said media file storing media content, said system comprising:

a user interface component of a media player for enabling a user of the first computing device to select the media file stored on the first computing device from which the item of initial metadata is identified, wherein the request is in response to the user selecting the media file;

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 on a second computing device storing additional metadata relating to the media content;

a weighting component for 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; and

a query component for searching the database for the additional metadata associated with the media content using the created token group and for providing the additional metadata from the database on the second computing device to the first computing device, wherein searching the database comprises generating a list of items of media content ordered according to the assigned weights of the extracted tokens, 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, and wherein the query component further performs additional searching based on the items corresponding to at least one of the other categories.

26. The system of claim 25, 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.

27. The system of claim 25, 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 first computing device.

28. The system of claim 25, wherein the system 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.

29. The system of claim 25, wherein the categories comprise at least one of the following:

album title, artist name, and song title.

30. A computer-readable storage medium of a first computing device having stored thereon a data structure describing an item of media content, said data structure comprising:

a plurality of metadata items including a metadata item associated with the media content;

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

a weight associated with each text string, said weight representing a frequency of occurrence of the text string in the metadata item associated with the media content; and

a token group including one or more of the tokens, wherein a second computing device stores the media content, and wherein the metadata item is provided to the second computing device from the first computing device responsive to a request from the second computing device as a function of comparing a token group identified in the request with the token group of said data structure, said token group identified in the request being associated with the media content stored by the second computing device, said media content stored in a media file selected by a user of the second computing device via a media player, wherein comparing a token group identified in the request with the token group of said data structure comprises generating a list of items of media content ordered according to the assigned weights, and wherein the request is in response to the user selecting the media file, wherein each of the tokens comprises a text string including a plurality of words, and further comprising a weight associated with each word of the text string, said weight representing a frequency of occurrence of the word of the text string in the metadata item.

31. The computer-readable storage medium of claim 30, wherein the token group comprises a permutation of a plurality of the tokens.

32. The computer-readable storage medium of claim 30, wherein the metadata item comprises one of the following:

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

33. A method of processing data via a first computing device for storage on a computer-readable storage medium, of the first computing device said method comprising:

receiving one or more metadata items at the first computing device 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;

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

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

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 the computer-readable medium of the first computing device to associate the media content with the stored metadata items, tokens, and token group, wherein a second computing device stores the media content, said media content stored in a media file selected by a user of the second computing device via a media player, and wherein at least one of the metadata items is provided from the first computing device to the second computing device responsive to a request from the second computing device as a function of comparing a token group identified in the request with the token group stored on the computer-readable medium of the first computing device, said token group identified in the request being associated with the media content stored by the second computing device, wherein the created token group comprises a text string and wherein comparing a token group identified in the request with the token group stored on the computer-readable medium of the first computing device comprises comparing a token group identified in the request with the token group stored on the computer-readable medium of the first computing device using a substring of less than the entire text string, and wherein the request is in response to the user selecting the media file.

34. The method of claim 33, 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.

35. The method of claim 33, 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.

36. The method of claim 35, wherein extracting one or more tokens further comprises omitting an extracted word based on its level of descriptiveness of the media content.

37. One or more computer-readable storage medium having computer-executable instructions for performing the method recited in claim 33.