Probabilities and reasoning about possibilities (Q757347)

This paper points out some properties that an inference system reasoning under uncertainty should have: calibration, audaciousness, and coherence. According with these properties the author presents some fundamental aspects of the relationship between probabilistic and symbolic reasoning. The principal result is that symbolic and quantitative approaches interact in interesting, sometimes counterintuitive ways. That it is counterproductive to use an uncertainty calculus to derive belief values instead of possibility ratios.