| CPC H04N 21/25883 (2013.01) [H04N 21/252 (2013.01); H04N 21/25891 (2013.01); H04N 21/812 (2013.01)] | 20 Claims |

|
1. A computer-implemented method for training a machine-learning model to predict television viewership, the computer-implemented method comprising:
receiving, by one or more processors, one or more historical viewership training profiles;
transforming, by the one or more processors, the one or more historical viewership training profiles to generate one or more multi-dimensional arrays;
training, by the one or more processors, a model based on the one or more multi-dimensional arrays to predict a viewership level forecast for each of one or more training profiles, wherein the training the model is based on one or more expected television viewership levels;
applying, by the one or more processors, the trained model to a second plurality of runtime television viewership profiles to generate a set of predicted runtime viewership characteristics for each of the second plurality of runtime television viewership profiles; and
storing, by the one or more processors, the set of predicted runtime viewership characteristics in one or more data stores.
|