Systems and methods for generating predictive data models using large data sets to provide personalized action recommendations

A computer-implemented method for generating a personalized action recommendation, comprising:

• acquiring a request for one or more conditions of a user;
• receiving user information from a plurality of search engines, wherein each of the plurality of search engines have different subset of the user information in different states;
• converting the received user information into a bundle by normalizing the received user information in the different states to a consistent format;
• generating features based on analysis of the bundle of normalized user information, including extrapolation based on stored medical history or derivation of standardized indices;
• generating one or more recommended actions for reaching a predetermined goal based on the generated features;
• generating an expected value for each of the one or more recommended actions based on applying a model to the normalized user information based on a first determination and applying the model to the normalized user information based on a second determination, wherein the first determination comprises the user performing the each of the one or more recommended actions and the second determination comprises the user not performing the each of the one or more recommended actions;
• ranking the one or more recommended actions based on the expected value for each of the one or more recommended actions;
• and
• determining an amount of resources for reaching the predetermined goal via the ranked one or more recommended actions.

A computer-implemented system for generating a personalized action recommendation, the system comprising:

• at least one non-transitory computer-readable medium configured to store instructions;
• and
• at least one processor configured to execute the instructions to cause the system to perform operations comprising:
• acquiring a request for one or more conditions of a user;
• receiving user information from a plurality of search engines, wherein each of the plurality of search engines have different subset of the user information in different states;
• converting the received user information into a bundle by normalizing the received user information in the different states to a consistent format;
• generating features based on analysis of the bundle of normalized user information, including extrapolation based on stored medical history or derivation of standardized indices;
• generating one or more recommended actions for reaching a predetermined goal based on the generated features;
• generating an expected value for each of the one or more recommended actions based on applying a model to the normalized user information based on a first determination and applying the model to the normalized user information based on a second determination, wherein the first determination comprises the user performing the each of the one or more recommended actions and the second determination comprises the user not performing the each of the one or more recommended actions;
• ranking the one or more recommended actions based on the expected value for each of the one or more recommended actions;
• and
• determining an amount of resources for reaching the predetermined goal via the ranked one or more recommended actions.