Incremental sequence completion system and method

• US 6,452,083 B2
• Filed: 07/02/2001
• Issued: 09/17/2002
• Est. Priority Date: 07/04/2000
• Status: Active Grant

• Current Assignee: Sony Europe (Amsterdam) BV (Sony Group Corp.)
• Original Assignee: Sony France SA (Sony Group Corp.)
• Inventors: Pachet, François, Cazaly, Daniel
• Primary Examiner(s): Fletcher, Marlon T.
• Application Number:
  US09/897,243
• Publication Number:
  US 20020002897A1
• Time in Patent Office: 442 Days
• Field of Search: 846/00, 846/01--02, 846/15, 846/09, 846/18, 846/26, 846/27, 846/49, 846/53, 846/56, 846/62--63
• US Class Current: 84/609
• CPC Class Codes: G06F 16/40 of multimedia data, e.g. slideshows comprising image and additional audio data retrieval of still image data G06F16/50; retrieval of audio data G06F16/60; retrieval of video data G06F16/70
• CPC Class Codes: G06F 16/637 Administration of user profiles, e.g. generation, initialization, adaptation or distribution
• 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: G10H 1/0058 Transmission between separate instruments or between individual components of a musical system G10H1/0083 takes precedence
• CPC Class Codes: G10H 2240/081 Genre classification, i.e. descriptive metadata for classification or selection of musical pieces according to style analysis genre G10H2210/036
• CPC Class Codes: G10H 2240/105 User profile, i.e. data about the user, e.g. for user settings or user preferences
• CPC Class Codes: H04N 21/4668 for recommending content, e.g. movies
• CPC Class Codes: H04N 21/8113 comprising music, e.g. song in MP3 format

Abstract

The present invention relates to a method designed to be used to compute music sequences in all possible contexts and situations, including: Internet adaptive radio, Digital Audio Broadcasting (DAB) with intelligent scheduling, music recommendation systems, and other innovative Electronic Music Distribution (EMD) services in general. These sequences are generated “iteratively”, and step by step. To the above end, there is provided a method of generating incrementally a sequence of items from a database containing said items, characterized in that the sequence is generated by implementing in combination a sequence completion system and a user profiling system, thereby taking into account both sequence coherence and user profile.

Claims

1. A method of selecting a desired next item in a sequence of items, comprising the steps of:

inputting a user profile corresponding to the preferences of a user of said sequence;

inputting control parameters for tuning the selection of said desired next item;

computing candidate items for said desired next item from a database of items, on the basis of the input user profile, the control parameters, and previous items in said sequence; and

selecting said desired next item from the computed candidate items by performing a collaborative filtering operation on the basis of the input user profile and an incremental sequence completion operation based on the input control parameters, thereby taking into account both sequence coherence and user profile.

2. The method according to claim 1, wherein each said item comprises at least one attribute.

3. The method according to claim 2, wherein that said items are linked to each other in said sequence by similarity relations in attributes of said items.

4. The method according to claim 1, wherein said sequence generating system is implemented by generating a desired next item in said sequence on the basis of similarity relationships between said item and said sequence.

5. The method according to claim 4, wherein said desired-next-item is further generated by user profiling techniques or metadata analysis techniques.

6. The method according to claim 1, wherein said sequence represents music titles.

7. The method according to claim 1, further comprising the step of providing said database with a parameter relating to a “

continuity”

mode on said sequence.

8. The method according to claim 1, further comprising the step of providing said database with a parameter relating to a “

repetitivity”

mode on said sequence.

9. The method according to claim 1, further comprising the step of providing said database with a parameter relating to a “

length of past”

mode on said sequence.

10. The method according to claim 1, further comprising the step of providing said database with a parameter relating to a “

explicit constraints”

mode on said sequence.

11. The method according to claim 1, further comprising the step of providing said database with a parameter relating to the “

number of items to be generated at a time”

mode.

12. The method according to claim 1, wherein said user profiling system is implemented using a parameter relating to a “

continuity”

mode on a user profile.

13. The method according to claim 1, wherein said database contains information representing a plurality of collections of descriptor and value pairs, each of said values for descriptors being selected from descriptor lists, and each of said descriptors is associated to a descriptor type.

14. The method according to claim 13, wherein said descriptor types at least comprise Integer-Type, Taxonomy-Type and Discrete-Type.

15. The method according to claim 14, wherein at least some of said descriptor types have mathematical similarity functions.

16. The method according to claim 1, wherein said database comprises musical pieces.

17. The method according to claim 1, wherein said sequence of items comprises music programs.

18. The method according to claim 2, wherein said database contains data corresponding to musical pieces and said attributes express objective data associated with a said item, such as a song title, author of said musical piece, duration of said musical piece, and recording label.

19. The method according to claim 2, wherein said database contains data corresponding to musical pieces and said attributes express subjective data, associated with a said item, which describe musical properties thereof, such as style, type of voice, music setup, type of instruments, tempo, type of melody, and main theme of the lyrics.

20. Implementation of the method according to claim 1 for creating a user recommendation system, each recommendation taking into account both sequence coherence and user profile.

21. Interactive radio station providing a personalised sequence of musical items, characterised in that said sequence is generated by a method according to claim 1, thereby taking into account user tastes interactively.

22. A system adapted to implement the method of claim 1, comprising a general-purpose computer and a monitor for display of the generated information.

23. A computer program product adapted to carry out the method of claim 1, when loaded into a general purpose computer.