Tomás E. Uribe, Vinay Chaudhri, Patrick J. Hayes, and Mark E. Stickel
Knowledge-based question answering relies on declarative knowledge and an inference procedure such as theorem proving. In this paper, we explore question answering based on spatial knowledge. We first consider a broad general-purpose axiomatic theory covering different aspects of qualitative spatial representation such as topology, orientation, distance, size, and shape. Since it can be expensive to build such a theory from scratch, we heuristically 81ice out a spatial subset of the Cyc knowledge base as a starting point for our work. We also explore a number of techniques to support efficient reasoning. The first is the RCC8 calculus, supported by the use of composition tables. We present a general-purpose mechanism for integrating composition tables into a first-order theorem prover. We also present a novel calculus to support reasoning with orientation. Finally, since a theoretical expressiveness analysis of such a broad spatial theory is not feasible, we develop a test suite of questions as a qualitative measure of the knowledge it captures. We also show how such a theory can serve as a special-purpose reasoning module in a larger system that is not limited to spatial queries.