Personalized auto-triage of communications

A system comprising:

• at least one processor;
• and
• a non-transitory processor-readable memory device storing instructions that when executed by the at least one processor causes the at least one processor to perform operations including:
• extracting natural language content from a piece of communication for a user;
• generating a representation of the piece of communication based on the natural language content extracted;
• generating, utilizing a global deep learning model, a first predicted degree of priority of the piece of communication from a perspective of multiple users based on the representation;
• generating, utilizing a personalized learning model for the user, a second predicted degree of priority of the piece of communication from both the perspective of the multiple users and an individual perspective of the user based on the first predicted degree of priority and user behavioral information associated with previous conversations of the user;
• and
• assigning a priority label to the piece of communication based on the second predicted degree of priority.

A computer program product comprising a non-transitory computer-readable storage medium having program code embodied therewith, the program code being executable by a computer to implement a method comprising:

• extracting natural language content from a piece of communication for a user;
• generating a representation of the piece of communication based on the natural language content extracted;
• generating, utilizing a global deep learning model, a first predicted degree of priority of the piece of communication from a perspective of multiple users based on the representation;
• generating, utilizing a personalized learning model for the user, a second predicted degree of priority of the piece of communication from both the perspective of the multiple users and an individual perspective of the user based on the first predicted degree of priority and user behavioral information associated with previous conversations of the user;
• and
• assigning a priority label to the piece of communication based on the second predicted degree of priority.