Conditional probabilistic reasoning without conditional logic
نویسنده
چکیده
Imaging is a class of non-Bayesian methods for the revision of probability density functions originally proposed as a semantics for conditional logic. Two of these revision functions, Standard Imaging and General Imaging, have successfully been applied to modelling information retrieval (IR). Due to the problematic nature of a“direct” implementation of Imaging revision functions, we propose their alternative implementation by representing the semantic structure that underlies them, in the language of a probabilistic (Bayesian) logic. Recasting these models of information retrieval in such a general-purpose knowledge representation (KR) tool, besides showing the potential of this “Bayesian” tool for the representation of non-Bayesian revision functions, paves the way to a possible integration of these models with other more KR-oriented models of IR, and to the exploitation of general purpose domain-knowledge.
منابع مشابه
Probabilistic Logic with Conditional Independence Formulae1
We investigate probabilistic propositional logic as a way of expressing and reasoning about uncertainty. In contrast to Bayesian networks, a logical approach can easily cope with incomplete information like probabilities that are missing or only known to lie in some interval. However, probabilistic propositional logic as described e.g. by Halpern [9], has no way of expressing conditional indepe...
متن کاملRepresenting Statistical Information and Degrees of Belief in First-Order Probabilistic Conditional Logic
Employing maximum entropy methods on probabilistic conditional logic has proven to be a useful approach for commonsense reasoning. Yet, the expressive power of this logic and similar formalisms is limited due to their foundations on propositional logic and in the past few years a lot of proposals have been made for probabilistic reasoning in relational settings. Most of these proposals rely on ...
متن کاملInverting Conditional Opinions in Subjective Logic
Subjective Logic has operators for conditional deduction and conditional abduction where subjective opinions are input arguments. With these operators traditional Bayesian reasoning can be generalised from taking only probabilistic arguments to also taking opinions as arguments, thereby allowing Bayesian modeling of situations affected by uncertainty and incomplete information. Conditional dedu...
متن کاملConditional Objects as Nonmonotonic Consequence Relationships
This paper is an investigation of the relationship between conditional objects obtained as a qualitative counterpart to conditional probabilities, and nonmonotonic reasoning. Roughly speaking, a conditional object can be seen as a generic rule which allows us to get a conclusion provided that the available information exactly corresponds to the "context" part of the conditional object. This giv...
متن کاملOn probabilistic inference in relational conditional logics
The principle of maximum entropy has proven to be a powerful approach for commonsense reasoning in probabilistic conditional logics on propositional languages. Due to this principle, reasoning is performed based on the unique model of a knowledge base that has maximum entropy. This kind of model-based inference fulfills many desirable properties for inductive inference mechanisms and is usually...
متن کامل