Peter Clark, John Thompson, and Bruce Porter
Our long-term research goal is to create systems capable of answering a wide variety of questions, including questions which were unanticipated at the time the system was constructed, and questions tailored to novel scenarios which the user is interested in. Our approach is to augment on-line text with a knowledgebased question-answering component, capable of reasoning about a scenario which the user provides, and synthesizing customized answers at run-time to his/her questions. To answer a question, the system creates a model of the user’s scenario of interest, infers the facts about it required for the answer, assembles those facts into a single structure, and presents it to the user. This process is guided by answer schemata, specifying what information should be included in the answer and how it should be organized. In this paper, we describe an implemented system based on this approach, which has been applied to several different application domains. We conclude with a discussion of the strengths and limitations of this approach.