Sentiment-based classification of media content

A non-transitory computer-readable storage medium having computer-executable code for determining a similarity value for a plurality of items of media content, the computer-executable code when executed causing steps to be performed comprising:

• identifying a set of comments associated with a first item of media content;
• generating a set of sentiment scores associated with the set of comments, wherein each sentiment score in the set of sentiment scores is associated with one of a plurality of different types of sentiment and each sentiment score indicates a type of sentiment expressed in a comment from the set of comments;
• adjusting a sentiment score from the set of sentiment scores associated with a comment provided by a user based on information describing sentiment expressed by the user in additional comments for additional media content items;
• combining sentiment scores in the set of sentiment scores to generate a plurality of aggregate sentiment scores, each of the aggregate sentiment scores indicating a magnitude of a different type of sentiment expressed in the set of comments;
• generating a first sentiment profile responsive at least in part to the plurality of aggregate sentiment scores, the first sentiment profile indicating different types of sentiment expressed in comments associated with the first item of media content;
• identifying a second sentiment profile for a second item of media content, the second sentiment profile indicating different types of sentiment expressed in comments associated with the second item of media content;
• comparing the first sentiment profile and the second sentiment profile;
• generating a similarity value responsive at least in part to the comparison, the similarity value indicating a measure of correspondence between one or more sentiments expressed in the comments associated with the first item of media content and one or more sentiments expressed in comments associated with the second item of media content;
• and
• responsive to the generated similarity value, transmitting instructions to display the second item of media content as a related item of the first item of media item.

A computer system for determining a similarity value for a plurality of items of media content, the system comprising:

• a computer processor;
• and
• a non-transitory computer-readable storage medium having computer-executable code that when executed by the processor causes steps to be performed comprising:
• identifying a set of comments associated with a first item of media content;
• generating a set of sentiment scores associated with the set of comments, wherein each sentiment score in the set of sentiment scores is associated with one of a plurality of different types of sentiment and each sentiment score indicates a type of sentiment expressed in a comment from the set of comments;
• adjusting a sentiment score from the set of sentiment scores associated with a comment provided by a user based on information describing sentiment expressed by the user in additional comments for additional media content items;
• combining sentiment scores in the set of sentiment scores to generate a plurality of aggregate sentiment scores, each of the aggregate sentiment scores indicating a magnitude of a different type of sentiment expressed in the set of comments;
• generating a first sentiment profile responsive at least in part to the plurality of aggregate sentiment scores, the first sentiment profile indicating different types of sentiment expressed in comments associated with the first item of media content;
• identifying a second sentiment profile for a second item of media content, the second sentiment profile indicating different types of sentiment expressed in comments associated with the second item of media content;
• comparing the first sentiment profile and the second sentiment profile;
• generating a similarity value responsive at least in part to the comparison, the similarity value indicating a measure of correspondence between one or more sentiments expressed in the comments associated with the first item of media content and one or more sentiments expressed in comments associated with the second item of media content;
• and
• responsive to the generated similarity value, transmitting instructions to display the second item of media content as a related item of the first item of media item.