Rules, Belief Functions and Default Logic

This paper describes a natural framework for rules , based on belief functions , which includes a repre­ s entation of numerical rules, default rules and rules allowing and rules not allowing contraposition. In particular it justifies the use of th e Dempster-Shafer Th eory for representing a particular class of rules , Belief calculated being a lower probability given cer­ tain independence assumptions on an underlying s pace. It s h ows h ow a belief function framework can be generalised to other logics , including a general Monte-Carlo algorithm for calculating belief, and h ow a version of Reiter's Default Logic can be seen as a limiting case of a belief function formalism.