www.clt.mq.edu.au/Events/Conferences/aaai05/
A Question Answering Architecture for Integrating Diverse Knowledge Resources
Boris Katz
Boris Katz is a Principal Research Scientist at the MIT Computer Science and Artificial
Intelligence Laboratory and head of the Laboratory's InfoLab
Group. His research interests include natural language generation and
understanding, machine learning and intelligent information access.
Over the last several years Boris Katz has been developing the START
natural language system that allows the user to access various
types of information using English.
Restricted domains offer important challenges and opportunities for the creation of question answering systems. A key benefit of restricted-domain systems is that they can provide high precision over their areas of coverage. But once we have created a large number of such systems, what next? Will users be required to choose among available QA systems, or will there be a supervisory QA system that helps route questions in appropriate directions? This talk presents a broad-focused architecture for integrating many such systems---restricted- and unrestricted-domain QA systems, multi-media repositories, arbitrary resources on the Web, and other information sources---under the coverage of a single, unifying QA interface. Key to this integration, by one view, is the use of language-motivated representations and techniques. Elements of such an approach will be illustrated in the context of particular aspects of the START question answering system.