Utilizing machine learning and composite utility scores from multiple event categories to improve digital content distribution

Abstract

The present disclosure is directed toward systems, methods, and non-transitory computer readable media for providing digital content to users by applying a machine learning model based on composite utility scores reflecting multiple events categories. For example, the systems described herein can identify, from a digital content publisher, significance ratings of various event categories that a user can perform. The systems can analyze user activities to determine a composite utility score for user based on events that the users have performed. Furthermore, in one or more embodiments, the systems train a machine learning model based on training composite utility scores to identify additional users likely to have elevated composite utility scores. Moreover, the disclosed systems can utilize the trained machine learning model to provide targeted digital content to computing devices of these additional users.

What is claimed is: 
     
         1 . A computer-implemented method comprising:
 determining a first significance of a first event category and a second significance of a second event category; and   generating a machine learning model for providing targeted digital content to computing devices of additional users from the first significance of the first event category and the second significance of the second event category by:
 identifying, from a digital repository of client device activity, a training user that has performed one or more events from the first event category and one or more events from the second event category, wherein the training user has a corresponding set of user characteristics; 
 generating a training composite utility score for the training user by applying the first significance to the one or more events from the first event category performed by the training user and the second significance to the one or more events from the second event category performed by the training user; 
 generating a predicted significance rating utilizing the machine learning model from the set of user characteristics; and 
 training the machine learning model by comparing the predicted significance rating to the training composite utility score. 
   
     
     
         2 . The computer-implemented method of  claim 1 , further comprising utilizing the machine learning model to provide digital content to a client device of an additional user. 
     
     
         3 . The computer-implemented method of  claim 2 , wherein utilizing the machine learning model to provide the digital content to the client device further comprises:
 identifying an additional set of user characteristics corresponding to the client device of the additional user;   generating a predicted composite utilize score utilizing the machine learning model from the additional set of user characteristics; and   providing the digital content to the client device based on the predicted composite utility score.   
     
     
         4 . The computer-implemented method of  claim 3 , wherein providing the digital content to the client device based on the predicted composite utility score comprises comparing the predicted composite utility score to a threshold composite utility score. 
     
     
         5 . The computer-implemented method of  claim 1 , further comprising:
 providing, to a publisher computing device, a user interface comprising one or more selectable significance options; and   identifying the first significance of the first event category and the second significance of the second event category based on user interaction with the one or more selectable significance options.   
     
     
         6 . The computer-implemented method of  claim 1 , further comprising determining the first significance of the first event category and the second significance of the second event category utilizing an additional machine learning model. 
     
     
         7 . The computer-implemented method of  claim 1 , wherein generating the training composite utility score comprises:
 determining a first event category score corresponding to the one or more events from the first event category performed by the training user and a second event category score corresponding to the one or more events from the second event category performed by the training user; and   applying a first weight corresponding to the first significance to the first event category score and a second weight corresponding to the second significance to the second event category score.   
     
     
         8 . The computer-implemented method of  claim 7 , further comprising normalizing the first event category score by comparing the first event category score associated with the training user with additional first event category scores associated with additional users. 
     
     
         9 . A system comprising:
 at least one processor; and   at least one non-transitory computer-readable storage medium storing instructions thereon that, when executed by the at least one processor, cause the system to:   determine a first significance of a first event category and a second significance of a second event category; and   generate a machine learning model for providing targeted digital content to computing devices of additional users from the first significance of the first event category and the second significance of the second event category by:
 identifying, from a digital repository of client device activity, a training user that has performed one or more events from the first event category and one or more events from the second event category, wherein the training user has a corresponding set of user characteristics; 
 generating a training composite utility score for the training user by applying the first significance to the one or more events from the first event category performed by the training user and the second significance to the one or more events from the second event category performed by the training user; 
 generating a predicted significance rating utilizing the machine learning model from the set of user characteristics; and 
 training the machine learning model by comparing the predicted significance rating to the training composite utility score. 
   
     
     
         10 . The system of  claim 9 , further comprising instructions that, when executed by the at least one processor, cause the system to utilize the machine learning model to provide digital content to a client device of an additional user by:
 identifying an additional set of user characteristics corresponding to the client device of the additional user;   generating a predicted composite utilize score utilizing the machine learning model from the additional set of user characteristics; and   providing the digital content to the client device based on the predicted composite utility score.   
     
     
         11 . The system of  claim 10 , wherein providing the digital content to the client device based on the predicted composite utility score comprises comparing the predicted composite utility score to a threshold composite utility score. 
     
     
         12 . The system of  claim 9 , further comprising instructions that, when executed by the at least one processor, cause the system to:
 provide, to a publisher computing device, a user interface comprising one or more selectable significance options; and   identify the first significance of the first event category and the second significance of the second event category based on user interaction with the one or more selectable significance options.   
     
     
         13 . The system of  claim 9 , further comprising instructions that, when executed by the at least one processor, cause the system to determine the first significance of the first event category and the second significance of the second event category utilizing an additional machine learning model. 
     
     
         14 . The system of  claim 9 , wherein generating the training composite utility score comprises:
 determining a first event category score corresponding to the one or more events from the first event category performed by the training user and a second event category score corresponding to the one or more events from the second event category performed by the training user; and   applying a first weight corresponding to the first significance to the first event category score and a second weight corresponding to the second significance to the second event category score.   
     
     
         15 . The system of  claim 14 , further comprising instructions that, when executed by the at least one processor, cause the system to normalize the first event category score by comparing the first event category score associated with the training user with additional first event category scores associated with additional users. 
     
     
         16 . A non-transitory computer-readable medium storing instructions that, when executed by at least one processor, cause a computing device to:
 determine a first significance of a first event category and a second significance of a second event category; and   generate a machine learning model for providing targeted digital content to computing devices of additional users from the first significance of the first event category and the second significance of the second event category by:
 identifying, from a digital repository of client device activity, a training user that has performed one or more events from the first event category and one or more events from the second event category, wherein the training user has a corresponding set of user characteristics; 
 generating a training composite utility score for the training user by applying the first significance to the one or more events from the first event category performed by the training user and the second significance to the one or more events from the second event category performed by the training user; 
 generating a predicted significance rating utilizing the machine learning model from the set of user characteristics; and 
 training the machine learning model by comparing the predicted significance rating to the training composite utility score. 
   
     
     
         17 . The non-transitory computer-readable medium of  claim 16 , further comprising instructions that, when executed by the at least one processor, cause the computing device to utilize the machine learning model to provide digital content to a client device of an additional user by:
 identifying an additional set of user characteristics corresponding to the client device of the additional user;   generating a predicted composite utilize score utilizing the machine learning model from the additional set of user characteristics; and   providing the digital content to the client device based on the predicted composite utility score.   
     
     
         18 . The non-transitory computer-readable medium of  claim 17 , wherein providing the digital content to the client device based on the predicted composite utility score comprises comparing the predicted composite utility score to a threshold composite utility score. 
     
     
         19 . The non-transitory computer-readable medium of  claim 16 , further comprising instructions that, when executed by the at least one processor, cause the computing device to:
 provide, to a publisher computing device, a user interface comprising one or more selectable significance options; and   identify the first significance of the first event category and the second significance of the second event category based on user interaction with the one or more selectable significance options.   
     
     
         20 . The non-transitory computer-readable medium of  claim 16 , further comprising instructions that, when executed by the at least one processor, cause the computing device to determine the first significance of the first event category and the second significance of the second event category utilizing an additional machine learning model.