A natural language interface can make a geographic information system (GIS)
easier to use. It also allows one to query a GIS using vague or imprecise
terms that are commonly used by end users. Major challenges in building a n
atural language interface include ambiguity and impreciseness. A natural la
nguage query may be ambiguous when it can be parsed in different ways. It m
ay contain imprecise terms that have to be processed before an executable q
uery can be generated. In this paper, we discuss the techniques we have dev
eloped, which are based on the possibility theory and fuzzy grammar theory,
for handling ambiguity and impreciseness in natural language queries. (C)
2000 Published by Elsevier Science B.V. All rights reserved.