Using fuzzy intervals to represent measurement error and scientificuncertainty in endangered species classification

摘要

Although fuzzy numbers (including fuzzy intervals) are often usedto capture semantic ambiguity, they are also useful to represent andpropagate measurement error. In this application, a classificationscheme used by international authorities for assigning biologicalspecies into categories of relative endangerment is generalized toaccept intervals and triangular or trapezoidal fuzzy numbers as inputsrepresenting empirical estimates of unknown quantities. Non-traditionaldefinitions for fuzzy magnitude comparisons and logical operations wererequired but, otherwise, standard fuzzy arithmetic was used. Adefuzzification step, which explicitly reveals the analyst's attitudesregarding evidence, can condense the result from the fuzzifiedclassification scheme to a single category. But this step is notrequired and may be counterproductive