Methods and apparatus to detect boring media

Abstract

Methods, apparatus, systems, and articles of manufacture are disclosed for the detection of boring media. An example apparatus includes at least one memory, instructions, and processor circuitry to correlate first boring media event determinations with a first portion of first media events, train a machine learning model based on the first boring media event determinations, the machine learning model to predict second boring media event determinations associated with a second portion of the first media events, and, in response to a number of the second boring media event determinations correctly predicted by the machine learning model satisfying a threshold, deploy the machine learning model to predict third boring media event determinations associated with second media events.

What is claimed is: 
     
         1 . An apparatus to detect boring media, the apparatus comprising:
 at least one memory;   instructions; and   processor circuitry to execute the instructions to:
 correlate first boring media event determinations with a first portion of first media events; 
 train a machine learning model based on the first boring media event determinations, the machine learning model to predict second boring media event determinations associated with a second portion of the first media events; and 
 in response to a number of the second boring media event determinations correctly predicted by the machine learning model satisfying a threshold, deploy the machine learning model to predict third boring media event determinations associated with second media events. 
   
     
     
         2 . The apparatus of  claim 1 , wherein the first boring media event determinations include a boring media determination, the first portion of the first media events include a third media event, and the processor circuitry is to:
 at least one of:
 determine that the third media event does not have audience measurement value based on the third media event being one of a static or dynamic station identification logo, a webcam feed, a test screen, or a blank screen; or 
 determine that the third media event is not associated with media monitoring information, the media monitoring information including at least one of a watermark or an identifier; 
   determine that the third media event is boring based on at least one of the third media event not having audience measurement value or not being associated with the media monitoring information; and   correlate the boring media determination and the third media event.   
     
     
         3 . The apparatus of  claim 1 , wherein the first media events include a third media event associated with audio data and video data, and the processor circuitry is to at least one of:
 determine one or more image temporal variation (ITV) values based on at least one of a red plane, a green plane, or a blue plane of a first portion of the video data;   determine one or more image chromatic information (ICI) values based on a histogram of (i) the at least one of the red plane, the green plane, or the blue plane of the first portion of the video data, or (ii) at least one of a red plane, a green plane, or a blue plane of the second portion of the video data;   determine one or more sound temporal domain flatness (STF) values based on an envelope associated with a third portion of the audio data; or   determine one or more sound harmonic richness (SHR) values based on a Fourier transform of at least one of the third portion or a fourth portion of the audio data.   
     
     
         4 . The apparatus of  claim 3 , further including one or more mass storage devices, and the processor circuitry is to:
 obtain the third media event from the one or more mass storage devices; and   store at least one of the one or more ITV values, the one or more ICI values, the one or more STF values, or the one or more SHR values in the one or more mass storage devices.   
     
     
         5 . The apparatus of  claim 1 , wherein the processor circuitry is to:
 in response to a first identification that a first one of the first media events is boring media, correlate a first label of boring media with the first one of the first media events; and   in response to a second identification that a second one of the first media events is not boring media, correlate a second label of not boring media with the second one of the first media events.   
     
     
         6 . The apparatus of  claim 1 , wherein the processor circuitry is to:
 correlate fourth boring media event determinations with the second portion of the first media events;   execute the machine learning model to predict the second boring media event determinations; and   determine the number of the second boring media event determinations that were correctly predicted based on comparisons of the second boring media event determinations and the fourth boring media event determinations.   
     
     
         7 . The apparatus of  claim 1 , wherein the processor circuitry is to retrain the machine learning model based on the first media events in response to a determination that the number of the second boring media event determinations that were correctly predicted do not satisfy the threshold. 
     
     
         8 . The apparatus of  claim 1 , wherein the number is a first number, the threshold is a first threshold, and the processor circuitry is to:
 store boring media events in a datastore; and   in response to a determination that a second number of the boring media events satisfies a second threshold, retrain the machine learning model based on the boring media events.   
     
     
         9 . An apparatus to detect boring media, the apparatus comprising:
 means for training a machine learning model, the means for training to:
 associate first boring media event determinations with a first portion of first media events; and 
 train the machine learning model based on the first boring media event determinations, the machine learning model to output second boring media event determinations associated with a second portion of the first media events; and 
   means for deploying the machine learning model to output third boring media event determinations associated with second media events, the means for deploying to deploy the machine learning model in response to a number of the second boring media event determinations correctly outputted by the machine learning model satisfying a threshold.   
     
     
         10 . The apparatus of  claim 9 , wherein the first boring media event determinations include a boring media determination, the first portion of the first media events include a third media event, and the means for training is to:
 at least one of:
 determine that the third media event does not have audience measurement value based on the third media event being one of a static or dynamic station identification logo, a webcam feed, a test screen, or a blank screen; or 
 determine that the third media event is not associated with media monitoring information, the media monitoring information including at least one of a watermark or an identifier; 
   determine that the third media event is boring based on at least one of the third media event not having audience measurement value or not being associated with the media monitoring information; and   correlate the boring media determination and the third media event.   
     
     
         11 . The apparatus of  claim 9 , wherein the first media events include a third media event associated with audio data and video data, and the means for training is to at least one of:
 generate one or more image temporal variation (ITV) values based on at least one of a red plane, a green plane, or a blue plane of a first portion of the video data;   generate one or more image chromatic information (ICI) values based on a histogram of (i) the at least one of the red plane, the green plane, or the blue plane of the first portion of the video data, or (ii) at least one of a red plane, a green plane, or a blue plane of the second portion of the video data;   generate one or more sound temporal domain flatness (STF) values based on an envelope associated with a third portion of the audio data; or   generate one or more sound harmonic richness (SHR) values based on a Fourier transform of at least one of the third portion or a fourth portion of the audio data.   
     
     
         12 . The apparatus of  claim 11 , further including means for storing, and the means for training is to:
 retrieve the third media event from the means for storing; and   store at least one of the one or more ITV values, the one or more ICI values, the one or more STF values, or the one or more SHR values in the means for storing.   
     
     
         13 . The apparatus of  claim 9 , wherein the means for training is to:
 in response to a first determination that a first one of the first media events is boring media, associate a first label of boring media with the first one of the first media events; and   in response to a second determination that a second one of the first media events is not boring media, associate a second label of not boring media with the second one of the first media events.   
     
     
         14 . The apparatus of  claim 9 , wherein the means for training is to:
 associate fourth boring media event determinations with the second portion of the first media events;   execute the machine learning model to output the second boring media event determinations; and   count the number of the second boring media event determinations that were correctly outputted based on comparisons of the second boring media event determinations and the fourth boring media event determinations.   
     
     
         15 . The apparatus of  claim 9 , wherein the means for training is to retrain the machine learning model based on the first media events in response to a determination that the number of the second boring media event determinations that were correctly outputted do not satisfy the threshold. 
     
     
         16 . The apparatus of  claim 9 , wherein the number is a first number, the threshold is a first threshold, and further including:
 means for storing boring media events; and   in response to a determination that a second number of the boring media events satisfy a second threshold, retrain the machine learning model based on the boring media events.   
     
     
         17 . At least one non-transitory computer readable storage medium comprising instructions that, when executed, cause processor circuitry to at least:
 generate associations of first boring media event determinations and a first portion of first media events;   train a machine learning model based on the first boring media event determinations, the machine learning model to determine second boring media event determinations associated with a second portion of the first media events; and   in response to a number of second boring media event determinations correctly determined by the machine learning model satisfying a threshold, cause execution of the machine learning model to determine third boring media event determinations associated with second media events.   
     
     
         18 . The at least one non-transitory computer readable storage medium of  claim 17 , wherein the first boring media event determinations include a boring media determination, the first portion of the first media events include a third media event, and the instructions, when executed, cause the processor circuitry to:
 at least one of:
 determine that the third media event does not have audience measurement value based on the third media event being one of a static or dynamic station identification logo, a webcam feed, a test screen, or a blank screen; or 
 determine that the third media event is not associated with media monitoring information, the media monitoring information including at least one of a watermark or an identifier; 
   determine that the third media event is boring based on at least one of the third media event not having audience measurement value or not being associated with the media monitoring information; and   correlate the boring media determination and the third media event.   
     
     
         19 . The at least one non-transitory computer readable storage medium of  claim 17 , wherein the first media events include a third media event associated with audio data and video data, and the instructions, when executed, cause the processor circuitry to at least one of:
 calculate one or more image temporal variation (ITV) values based on at least one of a red plane, a green plane, or a blue plane of a first portion of the video data;   calculate one or more image chromatic information (ICI) values based on a histogram of (i) the at least one of the red plane, the green plane, or the blue plane of the first portion of the video data, or (ii) at least one of a red plane, a green plane, or a blue plane of the second portion of the video data;   calculate one or more sound temporal domain flatness (STF) values based on an envelope associated with a third portion of the audio data; or   calculate one or more sound harmonic richness (SHR) values based on a Fourier transform of at least one of the third portion or a fourth portion of the audio data.   
     
     
         20 . The at least one non-transitory computer readable storage medium of  claim 17 , wherein the instructions, when executed, cause the processor circuitry to:
 in response to a first identification that a first one of the first media events is boring media, generate a first correlation of a first label of boring media with the first one of the first media events; and   in response to a second identification that a second one of the first media events is not boring media, generate a second correlation of a second label of not boring media with the second one of the first media events.   
     
     
         21 . The at least one non-transitory computer readable storage medium of  claim 17 , wherein the instructions, when executed, cause the processor circuitry to:
 associate fourth boring media event determinations with the second portion of the first media events;   invoke execution of the machine learning model to predict the second boring media event determinations; and   identify the number of the second boring media event determinations that were correctly determined based on comparisons of the second boring media event determinations and the fourth boring media event determinations.   
     
     
         22 . The at least one non-transitory computer readable storage medium of  claim 17 , wherein the instructions, when executed, cause the processor circuitry to retrain the machine learning model based on the first media events in response to a determination that the number of the second boring media event determinations that were correctly determined do not satisfy the threshold. 
     
     
         23 . The at least one non-transitory computer readable storage medium of  claim 17 , wherein the number is a first number, the threshold is a first threshold, and the instructions, when executed, cause the processor circuitry to:
 write boring media events in entries of a database; and   in response to a determination that a second number of the boring media events satisfy a second threshold, retrain the machine learning model based on the boring media events.