Clustering of musical content for playlist creation
AssigneeHarman International Industries, Incorporated
InventorRichard Allen Kreifeldt
Life of this patent
Estimating the term…
Abstract
One embodiment provides a non-transitory medium containing computer program code that performs an operation that includes determining an initial musical preferences model describing acoustical characteristics of musical content by identifying user preferences. Additionally, the operation includes dividing the initial musical preferences model into a plurality of cluster models based on the identified user preferences, each of the plurality of cluster models representing a respective set of acoustical characteristics of musical content. One of the plurality of cluster models is selected. The operation further includes creating a playlist by selecting instances of available musical content for playback using the selected cluster model.
Claims18
A non-transitory computer-readable medium containing instructions that, when executed by a processor, cause the processor to perform an operation comprising:
- classifying a user into a first affinity group included in a plurality of affinity groups based on metadata describing musical preferences of the user;
- generating, via the processor, an initial musical preferences model for the user based on:
- (i) the musical preferences of the user and (ii) a second musical preferences model that corresponds to the first affinity group;
- dividing, via the processor, the initial musical preferences model into a plurality of cluster models based on the musical preferences of the user, each of the plurality of cluster models representing a different set of acoustical characteristics of musical content;
- selecting one of the plurality of cluster models;
- and
- creating, via the processor, a playlist by selecting instances of available musical content for playback using the selected cluster model.
A system, comprising:
- a memory storing computer program code;
- and
- a processor that is coupled to the memory and, when executing the computer program code, performs an operation comprising:
- classifying a user into a first affinity group included in a plurality of affinity groups based on metadata describing musical preferences of the user;
- generating an initial musical preferences model for the user based on:
- (i) the musical preferences of the user and (ii) a second musical preferences model that corresponds to the first affinity group;
- dividing the initial musical preferences model into a plurality of cluster models based on the musical preferences of the user, each of the plurality of cluster models representing a different set of acoustical characteristics of musical content;
- selecting one of the plurality of cluster models;
- and
- creating a playlist by selecting instances of available musical content for playback using the selected cluster model.
A method, comprising:
- classifying a user into a first affinity group included in a plurality of affinity groups based on metadata describing musical preferences of the user;
- generating, via a processor, an initial musical preferences model for the user based on:
- (i) the musical preferences of the user and (ii) a second musical preferences model that corresponds to the first affinity group;
- dividing, via the processor, the initial musical preferences model into a plurality of cluster models based on the musical preferences of the user, each of the plurality of cluster models representing a different set of acoustical characteristics of musical content;
- selecting one of the plurality of cluster models;
- and
- creating, via the processor, a playlist by selecting instances of available musical content for playback using the selected cluster model.
Dependencies are inferred from the claim text.