System and method for providing crowd sourced metrics for network content broadcasters

A computer-implemented system, comprising:

• one or more processors;
• and
• a non-transitory computer readable storage medium including instructions that, when executed by the one or more processors, cause the one or more processors to perform operations including:
• dynamically obtaining broadcast content consumption information during playback of a broadcast stream associated with broadcast content, wherein the broadcast content consumption information corresponds to a user at an event that is associated with the broadcast content;
• identifying event content consumption information for a plurality of events taking place during a plurality of broadcast streams, wherein identifying the event content consumption information includes using metadata detection, ID tag detection, header detection, voice recognition, image analysis, motion detection, or signal detection;
• dynamically determining event metrics over at least a segment of the broadcast stream based on the broadcast content consumption information, wherein an the event metric includes an indication of a duration of time of the broadcast event in which a display of a device of the user is on;
• determining a commonality among one or more additional users, wherein the one or more additional users are associated with the broadcast content, and wherein the one or more additional users are likely to interact with additional broadcast content over a future period of time;
• executing a trained machine-learning model, the trained machine-learning model generates predictions of future event metrics for one or more events associated with additional broadcast content over the future period of time, wherein the trained machine-learning model uses scaling, aggregation, regression, standard deviation, summarization, categorization, or probability testing on the broadcast content consumption information to generate the predictions, wherein the future event metrics are based on the event metrics, and wherein the future event metrics are associated with the commonality among the user and the one or more additional users;
• and
• facilitating a transmission that includes the future event metrics, wherein when the future event metrics are received at a content provider, the future event metrics are used by the content provider to continually optimize quantified consumption of the additional broadcast content by the user and the one or more additional users.

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

• dynamically obtaining broadcast content consumption information during playback of a broadcast stream associated with broadcast content, wherein the broadcast content consumption information corresponds to a user at an event that is associated with the broadcast content;
• identifying event content consumption information for a plurality of events taking place during a plurality of broadcast streams, wherein identifying the event content consumption information includes using metadata detection, ID tag detection, header detection, voice recognition, image analysis, motion detection, or signal detection;
• dynamically determining event metrics over at least a segment of the broadcast stream based on the broadcast content consumption information, wherein an event metric includes an indication of a duration of time of the broadcast event in which a display of a device of the user is on;
• determining a commonality among one or more additional users, wherein the one or more additional users are associated with the broadcast content, and wherein the one or more additional users are likely to interact with the additional broadcast content over a future period of time;
• executing a trained machine-learning model, the trained machine-learning model generates predictions of future event metrics for one or more events associated with additional broadcast content over the future period of time, wherein the trained machine-learning model uses scaling, aggregation, regression, standard deviation, summarization, categorization, or probability testing on the broadcast content consumption information to generate the predictions, wherein the future event metrics are based on the event metrics, and wherein the future event metrics are associated with the commonality among the user and the one or more additional users;
• and
• facilitating a transmission that includes the future event metrics, wherein when the future event metrics are received at a content provider, the future event metrics are used by the content provider to continually optimize quantified consumption the additional broadcast content by the user and the one or more additional users.