Method and apparatus for intent handler performance manager

Abstract

The intent handler includes at least one machine learning model trained to generate an action for controlling the communications network based on an intent from a consumer. An intent satisfaction score is received and used to determine performance of the intent handler in configuring the communications network to carry out the intent. An update to the machine learning model of the intent handler (22) is determined based on the determined performance. The update may be a change in status of active or idle, updating the configuration or resource allocation, determining re-deployment or removal of machine learning model, retraining or adding a new machine learning model. A model lifecycle manager may be used. Configuration of the communications network may also be based on the received knowledge, such as other intent reports and KPI measurements and the intent handler report. The satisfaction score may be received through representational state transfer, REST, APIs to interface with a consumer, and may be an array a 0-10 score or an enumeration, based on trust, consistency, accuracy and first contact resolution rate.

Images