Data driven dialog management

A computing system comprising:

• at least one processor;
• and
• at least one memory comprising instructions that, when executed by the at least one processor, cause the computing system to:
• receive first input data corresponding to a first natural language input, the first input data associated with a user profile;
• determine history data representing an interaction history associated with the user profile;
• determine user satisfaction data associated with a user satisfaction with a dialog corresponding to the interaction history;
• generate first dialog state data including the first input data, the history data, and the user satisfaction data;
• process, using a first trained model, the first dialog state data to determine a first request for a first application to generate a response to the first natural language input, wherein the first trained model is trained to determine an application, from a set of applications, to perform processing with respect to the first natural language input;
• based on the first request, process, using a second trained model of the first application, the first dialog state data to generate first content data responsive to the first natural language input, wherein the second trained model is trained to use the first dialog state data to determine an action, from a set of actions associated with the first application, to be performed in response to the first natural language input;
• determine first output data corresponding to the first content second data;
• and
• causing presentation of the first output data.