Computer system for autonomous, self-learning customer interaction analysis and predictive service case resolution in digital support platforms

Abstract

A computer system for autonomously self-learning customer interaction analysis and predictive service case resolution in digital support platforms, comprising an interaction acquisition unit, a customer support state object creation unit, a self-learning service case correlation unit, a predictive service fault detection unit, an autonomous service action selection unit, and an output interface, characterized in that the customer support state object creation unit generates dynamic customer support state objects from multiple customer interaction data and platform events, wherein each customer support state object contains at least one customer reference, one interaction channel, one time validity information, one service context value, one urgency value, one sentiment deviation value, one problem probability value, one service fault indicator, one readiness to resolve value, one escalation status, and one trust value.and wherein the autonomous service action selection unit releases a service action to the output interface only if the service context value, problem probability value, urgency value, trust value, and escalation status satisfy a predefined resolution release condition.