A Statistical Viewpoint on the Theory of Evidence (Classic Reprint) - Softcover

Inhaltsangabe

How to read evidence and beliefs as a shared vote among experts. This primer explains a way to view the Dempster/Shafer theory of evidence as statistics of expert opinions, using Bayes’ updating to connect belief values with probabilities. It introduces belief, plausibility, and commonality, and shows how a simple, trackable set of numbers can summarize complex uncertainty. Read this edition to see how a theory of evidence maps to a practical, Bayesian viewpoint.

The book argues that combining evidence can be seen as updating probabilities over the product space of experts. It compares the traditional mass-based approach with an alternative that keeps probabilistic (logarithmic) opinions, offering a simpler computational path. By tying the ideas to what experts think, the text clarifies the foundations and the trade‑offs involved in modeling uncertainty.

• Learn how belief and plausibility relate to lower and upper probabilities in a concrete way.


• See how combining pieces of evidence mirrors Bayesian updates on labeled possibilities.


• Explore a streamlined formulation that reduces the number of parameters while tracking expert opinions.


• Understand differences between ambiguous knowledge and uncertain knowledge through opinion distributions.

Ideal for readers of statistics, decision theory, and knowledge engineering who want a clear, applied perspective on the theory of evidence.

Die Inhaltsangabe kann sich auf eine andere Ausgabe dieses Titels beziehen.