Smart advanced content retrieval

Abstract

Methods, systems, and apparatuses for implementing advanced content retrieval are described. Machine learning methods may be implemented so that a system may predict when a user device may experience network disconnections. The system may also predict the type of content one or more applications on the user device may seek to download during the network disconnection period. Neural networks may be trained based on user activity log data and may implement machine-learning techniques to determine user preferences and settings for advanced content retrieval. The system may predict when a user may want to download content in advance, the type of content the user may be interested in, anticipated network connectivity, and anticipated battery consumption. The system may then generate recommendations for the user device based on the predictions. If a user agrees with the recommendations, the system may obtain and cache the content.

1 . A computer implemented method, comprising:
 training, by a user device of a user, a neural network system using activity data stored locally on the user device;   by the user device and based on the training of the neural network system using the activity data:
 identifying a plurality of previous time periods in which one or more conditions were satisfied; 
 determining a first future time period when the one or more conditions will be satisfied, wherein the first future time period extends from a first particular future time to a second particular future time; and 
 generating a recommendation to schedule acquisition of electronic content data at the first particular future time if the one or more conditions are satisfied; 
   receiving, by the user device and prior to the first particular future time, input from the user indicating user acceptance of the recommendation to schedule acquisition of the electronic content data, the first particular future time recommended to download the electronic content data, and the one or more conditions recommended for acquiring the electronic content data at the first particular future time;   further training, by the user device, the neural network system based on the input received from the user;   determining, by the user device and using the further trained neural network system, a particular output time, during the first future time period and before the second particular future time, to output the electronic content data;   acquiring, by the user device, the electronic content data at the first particular future time in response to receiving the input from the user indicating the user acceptance; and   providing, by the user device, the electronic content data for display at the particular output time.   
     
     
         2 . The method of  claim 1 , wherein further training the neural network system based on the input received from the user includes further training the neural network system based on the input received from the user indicating the user acceptance of the recommendation to schedule acquisition of the electronic content data, the first particular future time recommended to download the electronic content data, and the one or more conditions recommended for acquiring the electronic content data at the first particular future time. 
     
     
         3 . The method of  claim 1 , wherein the plurality of previous time periods are identified based on determining that the user device experienced intermittent network connectivity during each of the plurality of previous time periods, and wherein the user device experiencing intermittent network connectivity is in addition to the one or more conditions. 
     
     
         4 . The method of  claim 1 , wherein providing the electronic content data for display at the particular output time is performed based on determining, during the first future time period and subsequent to determining the particular output time, that the one or more conditions are satisfied. 
     
     
         5 . The method of  claim 1 , wherein the activity data includes user data indicating user interactions occurring on the user device and user device data including sensor data associated with one or more sensors of the user device. 
     
     
         6 . The method of  claim 5 , wherein the one or more conditions include one or more particular types of user interactions occurring on the user device and/or one or more particular threshold levels associated with particular sensor data of the user device. 
     
     
         7 . The method of  claim 1 , wherein determining the first future time period includes determining the first particular future time based on predicting variations in network connectivity for the user device expected to occur during the first future time period while the electronic content data is being downloaded. 
     
     
         8 . The method of  claim 1 , wherein generating the recommendation includes determining, at a current time, a particular future time to provide the recommendation to the user based on predicting when a variation in network connectivity of the user device will occur during the first future time period, wherein the determined particular future time occurs before the first future time period, and wherein the input from the user is received responsive to providing the generated recommendation for display to the user at the determined particular future time. 
     
     
         9 . A system, comprising:
 one or more computers and one or more storage devices storing instructions that are operable and when executed by the one or more computers cause the one or more computers to perform operations comprising:
 training, by a user device of a user, a neural network system using activity data stored locally on the user device; 
 by the user device and based on the training of the neural network system using the activity data:
 identifying a plurality of previous time periods in which one or more conditions were satisfied; 
 determining a first future time period when the one or more conditions will be satisfied, wherein the first future time period extends from a first particular future time to a second particular future time; and 
 generating a recommendation to schedule acquisition of electronic content data at the first particular future time if the one or more conditions are satisfied; 
 
 receiving, by the user device and prior to the first particular future time, input from the user indicating user acceptance of the recommendation to schedule acquisition of the electronic content data, the first particular future time recommended to download the electronic content data, and the one or more conditions recommended for acquiring the electronic content data at the first particular future time; 
 further training, by the user device, the neural network system based on the input received from the user; 
 determining, by the user device and using the further trained neural network system, a particular output time, during the first future time period and before the second particular future time, to output the electronic content data; 
 acquiring, by the user device, the electronic content data at the first particular future time in response to receiving the input from the user indicating the user acceptance; and 
 providing, by the user device, the electronic content data for display at the particular output time. 
   
     
     
         10 . The system of  claim 9 , wherein further training the neural network system based on the input received from the user includes further training the neural network system based on the input received from the user indicating the user acceptance of the recommendation to schedule acquisition of the electronic content data, the first particular future time recommended to download the electronic content data, and the one or more conditions recommended for acquiring the electronic content data at the first particular future time. 
     
     
         11 . The system of  claim 9 , wherein the plurality of previous time periods are identified based on determining that the user device experienced intermittent network connectivity during each of the plurality of previous time periods, and wherein the user device experiencing intermittent network connectivity is in addition to the one or more conditions. 
     
     
         12 . The system of  claim 9 , wherein providing the electronic content data for display at the particular output time is performed based on determining, during the first future time period and subsequent to determining the particular output time, that the one or more conditions are satisfied. 
     
     
         13 . The system of  claim 9 , wherein the activity data includes user data indicating user interactions occurring on the user device and user device data including sensor data associated with one or more sensors of the user device. 
     
     
         14 . The system of  claim 13 , wherein the one or more conditions include one or more particular types of user interactions occurring on the user device and/or one or more particular threshold levels associated with particular sensor data of the user device. 
     
     
         15 . The system of  claim 9 , wherein determining the first future time period includes determining the first particular future time based on predicting variations in network connectivity for the user device expected to occur during the first future time period while the electronic content data is being downloaded. 
     
     
         16 . The system of  claim 9 , wherein generating the recommendation includes determining, at a current time, a particular future time to provide the recommendation to the user based on predicting when a variation in network connectivity of the user device will occur during the first future time period, wherein the determined particular future time occurs before the first future time period, and wherein the input from the user is received responsive to providing the generated recommendation for display to the user at the determined particular future time. 
     
     
         17 . One or more non-transitory computer-readable storage media comprising instructions, which, when executed by one or more processors, cause the one or more processors to perform operations comprising:
 training, by a user device of a user, a neural network system using activity data stored locally on the user device;   by the user device and based on the training of the neural network system using the activity data:
 identifying a plurality of previous time periods in which one or more conditions were satisfied; 
 determining a first future time period when the one or more conditions will be satisfied, wherein the first future time period extends from a first particular future time to a second particular future time; and 
 generating a recommendation to schedule acquisition of electronic content data at the first particular future time if the one or more conditions are satisfied; 
   receiving, by the user device and prior to the first particular future time, input from the user indicating user acceptance of the recommendation to schedule acquisition of the electronic content data, the first particular future time recommended to download the electronic content data, and the one or more conditions recommended for acquiring the electronic content data at the first particular future time;   further training, by the user device, the neural network system based on the input received from the user;   determining, by the user device and using the further trained neural network system, a particular output time, during the first future time period and before the second particular future time, to output the electronic content data;   acquiring, by the user device, the electronic content data at the first particular future time in response to receiving the input from the user indicating the user acceptance; and   providing, by the user device, the electronic content data for display at the particular output time.   
     
     
         18 . The one or more non-transitory computer-readable storage media of  claim 16 , wherein the plurality of previous time periods are identified based on determining that the user device experienced intermittent network connectivity during each of the plurality of previous time periods, and wherein the user device experiencing intermittent network connectivity is in addition to the one or more conditions. 
     
     
         19 . The one or more non-transitory computer-readable storage media of  claim 18 , wherein providing the electronic content data for display at the particular output time is performed based on determining, during the first future time period and subsequent to determining the particular output time, that the one or more conditions are satisfied. 
     
     
         20 . The one or more non-transitory computer-readable storage media of  claim 18 , wherein generating the recommendation includes determining, at a current time, a particular future time to provide the recommendation to the user based on predicting when a variation in network connectivity of the user device will occur during the first future time period, wherein the determined particular future time occurs before the first future time period, and wherein the input from the user is received responsive to providing the generated recommendation for display to the user at the determined particular future time.