Station library creation for a media service

A computing system comprising:

• at least one processor comprising microcircuits;
• non-transitory machine-readable storage;
• instructions stored in the non-transitory machine-readable storage for causing the at least one processor to carry out operations including:
• accessing collection metadata that describes media files included within a collection of media files;
• accessing seed metadata, the seed metadata being a basis on which a first subset of the collection of media files is to be defined;
• machine-generating, using a processor, a station set from the collection metadata based on the seed metadata, the machine-generated station set defining a station library by defining the first subset of the collection and referencing each media file in the first subset, wherein a boost value determined for each media file in the collection of media files augments a similarity score associated with the corresponding media file, wherein the boost value determined for each media file is based on at least one of (i) a relationship between an artist for the media file and a seed artist included in the collection metadata, or (ii) co-occurrences of the artist for the media file and the seed artist, wherein the corresponding media file is defined to be in the first subset of the collection based on the augmented similarity score;
• modifying the machine-generated station set based on a human-contributed input received from an editor device, the modified station set modifying the station library by defining a second subset of the collection and referencing each media file in the second subset;
• and
• configuring a media server to stream to a user device a media datastream that comprises media files selected from the second subset defined by the modified machine-generated station set, wherein the editor device comprises a first computer system, wherein the user device comprises a second computer system, wherein the user device is distinct from the editor device.

A non-transitory machine-readable storage medium comprising instructions that, when executed by one or more processors of a machine, cause the machine to perform operations comprising:

• accessing collection metadata that describes media files included within a collection of media files;
• accessing seed metadata, the seed metadata being a basis on which a first subset of the collection of media files is to be defined;
• machine-generating a station set from the collection metadata based on the seed metadata, the machine-generated station set defining a station library by defining the first subset of the collection and referencing each media file in the first subset, wherein a boost value determined for each media file in the collection of media files augments a similarity score associated with the corresponding media file, wherein the boost value determined for each media file is based on at least one of (i) a relationship between an artist for the media file and a seed artist included in the collection metadata, or (ii) co-occurrences of the artist for the media file and the seed artist, wherein the corresponding media file is defined to be in the first subset of the collection based on the augmented similarity score;
• modifying the machine-generated station set based on a human-contributed input received from an editor device, the modified station set modifying the station library by defining a second subset of the collection and referencing each media file in the second subset;
• and
• configuring a media server to stream to a user device a media datastream that comprises media files selected from the second subset defined by the modified machine-generated station set, wherein the editor device comprises a first computer system, wherein the user device comprises a second computer system, wherein the user device is distinct from the editor device.