Self-evolved adjustment framework for cloud-based large system based on machine learning

An information handling system comprising:

• one or more hardware processors;
• a memory coupled to at least one of the hardware processors;
• and
• a set of computer program instructions stored in the memory and executed by at least one of the hardware processors in order to perform actions comprising:
• training neural network, wherein the training comprises:
• receiving a plurality of data inputs to a cloud-based system from a plurality of users;
• gathering a plurality of sets of responsive system data from the cloud-based system, wherein each of the sets of responsive data corresponds to one or more of the received data inputs;
• and
• training the neural network by passing the received data inputs and corresponding sets of responsive data through the neural network;
• receiving one or more user actions from a user of the cloud-based system;
• inputting the one or more user actions to the trained neural network;
• receiving one or more predicted next user actions from the trained neural network;
• identifying that a selected one of the predicted next user actions is of a high importance and is outside of a pre-defined range;
• and
• performing a proactive response that is based on a behavior of the user that is identified as being outside of a pre-defined scope and the selected high importance predicted next user.

A computer program product stored in a computer readable storage medium, comprising computer program code that, when executed by an information handling system, performs actions comprising:

• training neural network, wherein the training comprises:
• receiving a plurality of data inputs to a cloud-based system from a plurality of users;
• gathering a plurality of sets of responsive system data from the cloud-based system, wherein each of the sets of responsive data corresponds to one or more of the received data inputs;
• and
• training the neural network by passing the received data inputs and corresponding sets of responsive data through the neural network;
• receiving one or more user actions from a user of the cloud-based system;
• inputting the one or more user actions to the trained neural network;
• receiving one or more predicted next user actions from the trained neural network;
• identifying that a selected one of the predicted next user actions is of a high importance and is outside of a pre-defined range;
• and
• performing a proactive response that is based on a behavior of the user that is identified as being outside of a pre-defined scope and the selected high importance predicted next user action.