Previous Chapter Chapter 25: Dempster-Shafer Theory Next Chapter
Chapter 25: Dempster-Shafer Theory
Dempster–Shafer theory (DST) is founded on the work of Dempster (1968) and Shafer (1976). Since its introduction the very name has caused confusion because it covers several models (Smets, 2002); a more general term often used is belief functions (both used intermittently here). Furthermore, DST has been considered a generalization (alternative) to the traditional Bayesian theory (Schubert, 1994). Alternatively, Shafer (1990) described belief functions as an alternative language of probability, not as something distinct from it.
As a general methodology, DST has formed the basis for the development of analysis techniques, including belief decision trees (Elouedi etal., 2001), DS/AHP (Beynon, 2002) and CaRBS (Beynon, 2005). In all these cases, they undertake analysis in the presence of ignorance. As such, DST is one of a ...