Auto playlist generator

• US 7,024,424 B1
• Filed: 03/11/2005
• Issued: 04/04/2006
• Est. Priority Date: 05/30/2001
• Status: Expired due to Term

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Platt, John C., Weare, Christopher B., Burges, Christopher, Zheng, Alice, Swnson, Steven E.
• Primary Examiner(s): ALAM, SHAHID AL
• Application Number:
  US11/077,646
• Time in Patent Office: 389 Days
• Field of Search: 707/101, 707/102, 707/104.1, 707/5, 707/6, 707/200, 715/501.1, 715/532, 706/12, 846/09
• US Class Current: 1/1
• CPC Class Codes: G06F 16/634 Query by example, e.g. query by humming
• CPC Class Codes: G06F 16/639 using playlists
• CPC Class Codes: G06F 16/68 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 16/683 using metadata automatically derived from the content
• CPC Class Codes: Y10S 707/916 Audio
• 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 707/99936 Pattern matching access
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing

Abstract

A system and method for generating a list is provided. The system includes a seed item input subsystem, an item identifying subsystem, a descriptive metadata similarity determining subsystem and a list generating subsystem that builds a list based, at least in part, on similarity processing performed on seed item descriptive metadata and user item descriptive metadata and user selected thresholds applied to such similarity processing. The method includes inexact matching between identifying metadata associated with new user items and identifying metadata stored in a reference metadata database. The method further includes subjecting candidate user items to similarity processing, where the degree to which the candidate user items are similar to the seed item is determined, and placing user items in a list of items based on user selected preferences for (dis)similarity between items in the list and the seed item.

Claims

1. A computer implemented method for generating a list, the computer implemented method comprising the following computer executable acts:

associating descriptive metadata with one or more candidate user items;

producing similarity data that characterizes the similarity between a candidate user item and at least one seed item;

producing a list of one or more user items related to the at least one seed item;

performing inexact matching between identifying metadata associated with a candidate user item to be added to a media library and identifying metadata associated with items in a reference metadata database; and

comparing descriptive metadata associated with the at least one seed item to descriptive metadata associated with the candidate user item, where comparing the descriptive metadata comprises;

comparing at least one feature vector associated with the at least one seed item to a feature vector associated with the candidate user item;

producing a difference vector related to the at least one feature vector associated with the at least one seed item and the feature vector associated with the candidate user item; and

producing the similarity data by employing the difference vector to retrieve a similarity value stored in a data store, where the data store was created by a machine learning technique.

2. The method of claim 1, the candidate user item is at least one of songs, music videos, movies, documents, books, and images.

3. The method of claim 1, where the candidate user item is a song, the candidate user item is identified by item identifying data, the item identifying data comprising at least one of a song artist, a song album and a song track.

4. The method of claim 3, where the feature vector associated with the candidate user item comprises fields associated with at least one of a genre feature, a subgenre feature, a style feature, a mood feature, a vocal coding feature, a rhythm type feature and a rhythm description feature.

5. A computer readable medium containing computer executable instructions for performing a method for generating a list, the method comprising the following computer related acts:

producing as-added descriptive metadata associated with at least one candidate user item to be added to a user media library;

producing similarity data that characterizes the similarity between the at least one candidate user item and at least one seed item;

producing a list of the at least one candidate user item related to the at least one seed item; and

performing inexact matching between identifying metadata associated with the at least one candidate user item to be added to a media library and identifying metadata associated with items in a reference metadata database; and

comparing descriptive metadata associated with the at least one seed item to as-added descriptive metadata associated with the at least one candidate user item, where comparing the descriptive metadata comprises;

comparing at least one feature vector associated with the at least one seed item to a feature vector associated with the at least one candidate user item;

producing a difference vector related to the at least one feature vector associated with the at least one seed item and the feature vector associated with the at least one candidate user item; and

producing the similarity data by employing the difference vector to retrieve a similarity value stored in a data store, where the data store was created by a machine learning technique.

6. The computer readable medium of claim 5, where the at least one candidate user item is at least one of songs, music videos, movies, documents, books, and images.

7. The computer readable medium of claim 5, further comprising associating descriptive metadata to the at least one candidate user item.

8. The computer readable medium of claim 7, associating descriptive metadata to the at least one candidate user item comprising performing inexact matching between identifying metadata associated with the at least one candidate user item and identifying metadata associated with items in a reference metadata database.

9. The computer readable medium of claim 5, the at least one feature vector associated with the at least one seed item comprising descriptive metadata relating to the seed item.

10. The computer readable medium of claim 5, the at least one feature vector associated with the at least one candidate user item comprising descriptive metadata relating to the candidate user item.

11. A computer implemented method for generating a list comprising the following computer executable acts:

associating descriptive metadata with at least one candidate user item;

producing similarity data that characterizes a similarity between the at least one candidate user item and at least one seed item; and

producing a list of the at least one candidate user item related to the at least one seed item; and

performing inexact matching between identifying metadata associated with the at least one candidate user item to be added to a media library and identifying metadata associated with items in a reference metadata database; and

comparing descriptive metadata associated with the at least one seed item to descriptive metadata associated with the at least one candidate user item, where comparing the descriptive metadata comprises;

comparing at least one feature vector associated with the at least one seed item to a feature vector associated with the at least one candidate user item;

producing a difference vector related to the at least one feature vector associated with the at least one seed item and the feature vector associated with the at least one candidate user item; and

producing the similarity data by employing the difference vector to retrieve a similarity value stored in a data store, where the data store was created by a machine learning technique.

12. The method of claim 11, the at least one candidate user item is at least one of songs, music videos, movies, documents, books, and images.

13. The method of claim 11, where the at least one candidate user item is a song, the at least one candidate user item is identified by item identifying data, the item identifying data comprising at least one of a song artist, a song album and a song track.

14. The method of claim 13, where the feature vector associated with the candidate user item comprises fields associated with at least one of a genre feature, a subgenre feature, a style feature, a mood feature, a vocal coding feature, a rhythm type feature and a rhythm description feature.