A Personalized Assistant for Customer Complaints Management Systems

Boris Galitsky

We build a personalized conflict resolution agent that applies reasoning about mental attributes to processing of scenarios of multiagent interactions. Our approach is deployed in the domain of complaint analysis: rather advanced user interface and machine learning are required to advise a customer on how to prepare a valid complaint and to assist in its formal structured representation. We demonstrate that exchanging information on mental actions of conflicting agents only without domain-specific knowledge is frequently sufficient to determine a complaint validity.


This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.