Inference with Possibilistic Evidence
نویسندگان
چکیده
In this paper, the concept of possibilistic ev idence which is a possibility distribution as well as a body of evidence is proposed over an infinite universe of discourse. The in ference with possibilistic evidence is investi gated based on a unified inference framework maintaining both the compatibility of con cepts and the consistency of the probability logic.
منابع مشابه
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 first is seen in connection with probabilistic, the second in connection with possibilistic reasoning, although we argue that both schemes are appl...
متن کاملExtending Description Logics with Uncertainty Reasoning in Possibilistic Logic
Possibilistic logic provides a convenient tool for dealing with inconsistency and handling uncertainty. In this paper, we propose possibilistic description logics as an extension of description logics. We give semantics and syntax of possibilistic description logics. We then define two inference services in possibilistic description logics. Since possibilistic inference suffers from the drownin...
متن کاملF 1 . 2 Inference Methods
This section investigates graphical modeling as a powerful framework for drawing inferences under imprecision and uncertainty. We survey the semantical background and relevant properties of relational, probabilistic, and possibilistic networks and consider evidence propagation in such networks as well as methods for learning them from data. Whereas the probabilistic Bayesian networks and Markov...
متن کاملLearning Possibilistic Networks with a Global Evaluation Method
Inference networks, probabilistic as well as possibilistic, are popular techniques to make reasoning in multi-dimensional domains feasible. Since constructing them by hand can be tedious and time consuming, a large part of recent research has been devoted to learning inference networks from data. Most of the proposed methods are based on local, i.e. single hyperedge evaluation. In this paper we...
متن کاملHybrid Possibilistic Networks
Possibilistic networks are important tools for dealing with uncertain pieces of information. For multiplyconnected networks, it is well known that the inference process is a hard problem. This paper studies a new representation of possibilistic networks, called hybrid possibilistic networks. The uncertainty is no longer represented by local conditional possibility distributions, but by their co...
متن کامل