Linear programming consumption learning system

A method for operating a digital content recommendation system, comprising:

• detecting and identifying user actions as events on network connected digital devices;
• relating the user actions to digital content;
• operating a timer to control a minimum time period filter on the user actions;
• operating a sequencer in conjunction with the timer, the minimum time period filter, and a combiner to transform the user actions into aggregate user actions;
• operating the content recommendation system by applying the aggregate user actions to evolve a computer-stored user profile according to the aggregate user actions;
• applying the user profile to select and route digital content to particular end user devices;
• operating timer override on the events, such that events that affect machine system learning are associated with an override time other than a current time output by the timer;
• and
• for a timeslot subscriber, applying the override time to determine which timeslot the event will influence the machine system learning, for viewed content that starts on a timeslot boundary.

A non-transitory computer-readable storage medium for operating a digital content recommendation system, the computer-readable storage medium including instructions that when executed by a computer, cause the computer to:

• detect and identify user actions on network connected digital devices;
• relate the user actions to digital content;
• operate a timer to control a minimum time period filter on the user actions;
• operate a sequencer in conjunction with the timer, the minimum time period filter, and a combiner to transform the user actions into aggregate user actions;
• operate the content recommendation system by applying the aggregate user actions to evolve a computer-stored user profile according to the aggregate user actions;
• applying the user profile to select and route digital content to particular end user devices;
• and
• operating timer override on the events, such that events that affect machine system learning are associated with an override time other than a current time output by the timer;
• and
• for a timeslot subscriber, applying the override time to determine which timeslot the event will influence the machine system learning, for viewed content that starts on a timeslot boundary.