Conservative, Proportional and Optimistic Contextual Discounting in the Belief Functions Theory

摘要

Information discounting plays an important role in the theory of belieffunctions and, generally, in information fusion. Nevertheless, neitherclassical uniform discounting nor contextual cannot model certain use cases,notably temporal discounting. In this article, new contextual discountingschemes, conservative, proportional and optimistic, are proposed. Someproperties of these discounting operations are examined. Classical discountingis shown to be a special case of these schemes. Two motivating cases arediscussed: modelling of source reliability and application to temporaldiscounting.