Fuzzy sets, uncertainty, and information
Fuzzy sets, uncertainty, and information
A possibilistic approach to qualitative model-based diagnosis
Telematics and Informatics - Special issue: artificial intelligence and advanced computing technologies for space applications
Possibilistic processes for complex systems modeling
Possibilistic processes for complex systems modeling
Mathematics of Data Fusion
Hi-index | 0.00 |
Most of the applications chapters in this book deal with the quantification of various forms of uncertainty, both numeric and nonnumeric. Uncertainty in numerical quantities can be random in nature, where probability theory is very useful, or it can be the result of bias or an unknown error, in which case fuzzy set theory, evidence theory, or possibility theory might prove useful. Probability theory also has been used almost exclusively to deal with the form of uncertainty due to chance (randomness), sometimes called variability, and with uncertainties arising from eliciting and analyzing expert information. Three other prevalent forms of uncertainty are those arising from ambiguity, vagueness, and imprecision. While vagueness and ambiguity can arise from linguistic uncertainty, they also can be associated with some numerical quantities, such as "approximately 5." Imprecision is generally associated with numerical quantities, although applications may exist where this type of uncertainty is nonnumeric (i.e., qualitative), for example, "the missile was close to the target." How do variability, ambiguity, vagueness, and imprecision differ as forms of uncertainty?In the sections that follow, we detail the most popular methods used to address these various forms of uncertainty whether they are quantitative or qualitative. While fuzzy set theory and probability theory have been used for all these forms of uncertainty, this chapter will extend this scope somewhat by commenting on the use of possibility theory in the characterization of ambiguity.