Systems and methods for guided user actions

Abstract

Systems and methods for guided user actions are described, including detecting a first action performed by a user, gathering information associated with the first action; retrieving a predictive model based on the information, determining an applicability level of the predictive model to the first action, the predictive model suggests a second action; providing the second action in a user interface when the applicability level meets a threshold level; and receiving input from the user selecting the second action or a third action.

What is claimed is: 
     
         1 . A method comprising:
 receiving, at data processing hardware, an indicator that a user consents to an automated machine learning process;   after receipt of the indicator that the user consents to the automated machine learning process, receiving, at data processing hardware, a history of past user data from the user;   generating, by the data processing hardware, a plurality of predictive models using machine learning based on the history of past user data from the user, the plurality of predictive models comprising a first predictive model and a second predictive model, the first predictive model relevant to a first type of data present in the history of the past user data from the user, the second predictive model relevant to a second type of data present in the history of the past user data from the user;   for each predictive model of the plurality of predictive models, determining, by the data processing hardware, whether a performance level of the respective predictive model satisfies a threshold level when the respective predictive model generates a prediction in response to input data, the input data corresponding to the respective type of data present in the history of the past user data from the user that generated the respective predictive model; and   when the first predictive model and the second predictive model satisfy the threshold level, selecting, by the data processing hardware, at least one of the first predictive model or the second predictive model to apply to data corresponding to a user input, the data corresponding to the user input being the first type of data or the second type of data.   
     
     
         2 . The method of  claim 1 , wherein the user input comprises a user action 
     
     
         3 . The method of  claim 2 , wherein the selected at least one of the first predictive model or the second predictive model generates a prediction for a subsequent user action. 
     
     
         4 . The method of  claim 3 , wherein the prediction is automatically executed on behalf of the user. 
     
     
         5 . The method of  claim 3 , wherein the user action comprises browsing one or more online sites in an open mode, and the subsequent user action comprises a user preference to browse in a private mode. 
     
     
         6 . The method of  claim 3 , wherein the user action comprises identifying content and the subsequent user action comprises selecting an application to process the identified content. 
     
     
         7 . The method of  claim 3 , wherein the user action comprises accessing a website and the subsequent user action comprises selecting a mode of a browser to access the website. 
     
     
         8 . The method of  claim 1 , wherein:
 the first type of data present in the history of the past user data from the user comprises data corresponding to web browsing activity; and   the second type of data present in the history of the past user data from the user comprises data corresponding to sharing activity on an online social circle of a social network.   
     
     
         9 . The method of  claim 1 , wherein:
 the first type of data present in the history of the past user data from the user comprises data corresponding to email communication; and   the second type of data present in the history of the past user data from the user comprises data corresponding to sharing activity on an online social circle of a social network.   
     
     
         10 . The method of  claim 1 , wherein determining whether a performance level of the respective predictive model satisfies a threshold level when the respective predictive model generates a prediction in response to input data comprises:
 performing a weighted calculation of attributes associated with the respective predictive model; and   determining a combination of at least some of the weighted calculation of the attributes meet the threshold level.   
     
     
         11 . A system comprising:
 data processing hardware; and   memory hardware in communication with the data processing hardware, the memory hardware storing instructions that when executed on the data processing hardware cause the data processing hardware to perform operations comprising:
 receiving an indicator that a user consents to an automated machine learning process; 
 after receipt of the indicator that the user consents to the automated machine learning process, receiving a history of past user data from the user; 
 generating a plurality of predictive models using machine learning based on the history of past user data from the user, the plurality of predictive models comprising a first predictive model and a second predictive model, the first predictive model relevant to a first type of data present in the history of the past user data from the user, the second predictive model relevant to a second type of data present in the history of the past user data from the user; 
 for each predictive model of the plurality of predictive models, determining whether a performance level of the respective predictive model satisfies a threshold level when the respective predictive model generates a prediction in response to input data, the input data corresponding to the respective type of data present in the history of the past user data from the user that generated the respective predictive model; and 
 when the first predictive model and the second predictive model satisfy the threshold level, selecting at least one of the first predictive model or the second predictive model to apply to data corresponding to a user input, the data corresponding to the user input being the first type of data or the second type of data. 
   
     
     
         12 . The system of  claim 11 , wherein the user input comprises a user action 
     
     
         13 . The system of  claim 12 , wherein the selected at least one of the first predictive model or the second predictive model generates a prediction fora subsequent user action. 
     
     
         14 . The system of  claim 13 , wherein the prediction is automatically executed on behalf of the user. 
     
     
         15 . The system of  claim 13 , wherein the user action comprises browsing one or more online sites in an open mode, and the subsequent user action comprises a user preference to browse in a private mode. 
     
     
         16 . The system of  claim 13 , wherein the user action comprises identifying content and the subsequent user action comprises selecting an application to process the identified content. 
     
     
         17 . The system of  claim 13 , wherein the user action comprises accessing a website and the subsequent user action comprises selecting a mode of a browser to access the website. 
     
     
         18 . The system of  claim 11 , wherein:
 the first type of data present in the history of the past user data from the user comprises data corresponding to web browsing activity; and   the second type of data present in the history of the past user data from the user comprises data corresponding to sharing activity on an online social circle of a social network.   
     
     
         19 . The system of  claim 11 , wherein:
 the first type of data present in the history of the past user data from the user comprises data corresponding to email communication; and   the second type of data present in the history of the past user data from the user comprises data corresponding to sharing activity on an online social circle of a social network.   
     
     
         20 . The system of  claim 11 , wherein determining whether a performance level of the respective predictive model satisfies a threshold level when the respective predictive model generates a prediction in response to input data comprises:
 performing a weighted calculation of attributes associated with the respective predictive model; and   determining a combination of at least some of the weighted calculation of the attributes meet the threshold level.