An alternative interpretation of linguistic variable is introduced together
with the notion of a linguistic description of a value or set of values. T
he latter is taken to be a fuzzy set on words where the membership values q
uantify the suitability of a particular word as a label for the value or va
lues being considered. This concept is then applied to reasoning with lingu
istic quantifiers these being defined as linguistic descriptions of probabi
lity values. From this viewpoint linguistic quantifiers are constraints on
probability values and hence using the voting model and Bayesian methods in
fer second order densities. In this respect such quantifiers can be view as
an alternative form of imprecise probability. These ideas are then used in
our proposed methodology for converting probabilistic inference rules into
linguistic inference rules and a computationally cheap approximation algor
ithm for such rules is then introduced. The approach is illustrated in a nu
mber of worked examples using various types of rules including linguistic s
yllogisms and a linguistic version of Jeffrey's rule. Finally a number of m
ethods for information fusion with linguistically quantified statements are
discussed. (C) 2001 Elsevier Science Inc. All rights reserved.