Enhancing media content effectiveness using feedback between evaluation and content editing

A system for improving media content effectiveness, the system comprising one or more processors configured to:

• generate an initial intermediate representation (IR) of a provided media content, wherein the initial IR specifies one or more editable elements of the provided media content;
• apply edits to the initial IR to generate candidate IR variations;
• execute a content creation application to create candidate media contents based on the candidate IR variations, wherein each candidate IR variation results in creation of a corresponding candidate media content;
• execute a content evaluation application to generate an effectiveness score for each candidate media content;
• prune the candidate IR variations to retain a threshold number of candidate IR variations as surviving IR variations;
• make a determination that one or more of (i) the effectiveness scores generated by the content evaluation application exceeds a first threshold value, or (ii) an improvement in at least one of the effectiveness scores relative to a previous iteration is less than a second threshold value;
• and
• in response to making the determination, execute the content creation application to create a modified media content based on one of the surviving IR variations.

A computer program product including one or more non-transitory machine-readable mediums encoded with instructions that when executed by one or more processors cause a process to be carried out for improving media content effectiveness, the process comprising:

• generating an initial intermediate representation (IR) of a provided media content, wherein the initial IR specifies one or more editable elements of the provided media content;
• applying edits to the initial IR to generate candidate IR variations;
• making a first determination that a particular one of the candidate IR variations generated by applying the edits is included in an IR variation history maintained in the one or more non-transitory machine-readable mediums;
• in response to making the first determination, removing the particular candidate IR variation from further consideration;
• executing a content creation application to create candidate media contents based on the candidate IR variations, wherein each candidate IR variation results in creation of a corresponding candidate media content, and wherein no additional candidate media content is created based on the particular candidate IR variation that was removed from further consideration;
• executing a content evaluation application to generate an effectiveness score for each candidate media content;
• and
• pruning the candidate IR variations to retain a threshold number of candidate IR variations as surviving IR variations.