Dynamic classification of attribute relevance and classification

A computer-implemented method comprising:

• in response to receiving, at a computer system, information about a keyword associated with a media work, identifying, by the computer system, a set of media works based at least in part on:
• aggregated classification information for a plurality of keywords associated with a plurality of media works, the aggregated classification information including one or more terms and representing a relationship of each keyword and term pair to each media work of the plurality of media works, the aggregated classification information determined based at least in part on first input from a plurality of users;
• aggregated rating information for the aggregated classification information, the aggregated rating information of said each keyword representing a relevancy of a particular keyword classification and at least one media work of the plurality of media works, the aggregated rating information determined based at least in part on second input from the plurality of users;
• and
• generating, by the computer system, ranking information for the plurality of keywords associated with each media work of the plurality of media works based at least in part on the aggregated classification information and the aggregated rating information.

A system, comprising:

• a processor;
• and
• memory including computer-executable instructions that, when executed by the processor, cause the system to at least:
• receive, from a plurality of users, classification information including one or more terms, the classification information representing a relationship of a keyword, of a plurality of keywords, and term pair to a media work;
• receive, from the plurality of users, rating information for a set of keywords associated with a set of media works, the rating information for each keyword associated with each media work representing a relevancy of a keyword and associated classification information with said media work;
• receive a recommendation request from a user, the recommendation request including information about an associated keyword of a particular media work;
• and
• provide for presentation to the user a plurality of media works based at least in part on the rating information and the classification information for the set of keywords for said each media work of the set of media works and the recommendation request.

A non-transitory computer-readable storage medium collectively storing computer-executable instructions that, when executed by a computer system, configure the computer system to collectively perform operations comprising:

• receiving information about a first media work and a second media work;
• accessing classification information and rating information for a set of keywords associated with the first media work and the second media work, the classification information including one or more terms, the classification information representing a relationship of each keyword, of the set of keywords, and a term pair that is associated with the first and second media works, the rating information representing a relevancy of the classification information and said each keyword with the first and second media works;
• providing information about additional media works to a user based at least in part on an algorithm utilizing the classification information and the rating information for the first media work and the second media work;
• and
• generating rank information for the set of keywords associated with the first media work and the second media work based at least in part on the classification information and the rating information for the first media work and the second media work.