Title: Intensional FOL for reasoning about probabilities and probabilistic logic programming
Authors: Zoran Majkić; Bhanu Prasad
ISRST, Roma, Italy
Department of Computer and Information Sciences, Florida A&M University, Tallahassee, Florida 32307, USA
Abstract: It is important to have a logic, both for computation of probabilities and for reasoning about probabilities, with well-defined syntax and semantics. The current approaches, which are based on Nilsson's probability structures/logics as well as linear inequalities, to reason about probabilities, have some deficiencies. In this research, we have presented a complete revision of those approaches and have shown that the logic for reasoning about probabilities can be naturally embedded into a 2-valued intensional first-order logic (FOL) with intensional abstraction, by avoiding current ad-hoc system composed of two different 2-valued logics: one for the classical propositional logic at a lower-level and a new one, at a higher-level, for probabilistic constraints with probabilistic variables. The theoretical results that are obtained are applied to probabilistic logic programming.
Keywords: probabilities; 2-valued intensional first-order logic; Nilsson's probability structures; linear inequalities.
Int. J. of Intelligent Information and Database Systems, 2018 Vol.11, No.1, pp.79 - 96
Submission date: 08 Nov 2017
Date of acceptance: 30 Nov 2017
Available online: 26 Apr 2018