Evaluation measures for learning probabilistic and possibilistic networks

Evidence propagation in inference networks, probabilistic or possibilistic, can be done in two different ways - using a product/sum scheme or using a minimum/maximum scheme - depending on the type of answers one expects from the network. Usually the former is seen in connection with probabilistic reasoning, and the latter with possibilistic reasoning, although we argue that both schemes are applicable in both settings. The paper discusses learning inference networks from data and examines some evaluation measures with respect to the chosen propagation method