Explaining Question Answering Systems with Contexts

Deborah L. McGuinness

One central issue in the usability of answers is in their understandability. We have focused a line of research on explaining answers from heterogeneous distributed systems with the goal of improving usability and trustability in answers by increasing answer understandability and trustability. We explore ways of providing an interoperable distributed infrastructure that supports explanations containing provenance concerning answers — where information came from, how reliable it is -- along with information about assumptions relied on, information manipulation techniques, etc. The infrastructure utilizes the Proof Markup Language, encoded in the OWL Web Ontology Language so as to be compatible with web standards and also so as to be able to leverage existing web ontologies. In this paper, we expose and explore some issues related to context as it is used in some of our question answering systems.

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.