A syntax-based framework for merging imprecise probabilistic logic programs

Authors:
Anbu Yue;Weiru Liu
Affiliations:
School of Electronics, Electrical Engineering and Computer Science, Queen's University Belfast, Belfast, UK;School of Electronics, Electrical Engineering and Computer Science, Queen's University Belfast, Belfast, UK
Venue:
IJCAI'09 Proceedings of the 21st international jont conference on Artifical intelligence
Year:
2009

Citing 8
Cited 1

Aggregating point estimates: a flexible modeling approach

Management Science
Methods for combining experts' probability assessments

Neural Computation
Probabilistic logic programming with conditional constraints

ACM Transactions on Computational Logic (TOCL)
DA2 merging operators

Artificial Intelligence - Special issue on nonmonotonic reasoning
Information markets vs. opinion pools: an empirical comparison

Proceedings of the 6th ACM conference on Electronic commerce
Belief revision and information fusion on optimum entropy: Research Articles

International Journal of Intelligent Systems - Uncertain Reasoning (Part 2)
Nonmonotonic probabilistic logics under variable-strength inheritance with overriding: Complexity, algorithms, and implementation

International Journal of Approximate Reasoning
Necessary and sufficient consensus conditions for the eventwise aggregation of lower probabilities

Fuzzy Sets and Systems

A logic with imprecise probabilities and an application to automated reasoning using rewriting techniques

Fuzzy Sets and Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we address the problem of merging multiple imprecise probabilistic beliefs represented as Probabilistic Logic Programs (PLPs) obtained from multiple sources. Beliefs in each PLP are modeled as conditional events attached with probability bounds. The major task of syntax-based merging is to obtain the most rational probability bound for each conditional event from the original PLPs to form a new PLP. We require the minimal change principle to be followed so that each source gives up its beliefs as little as possible. Some instantiated merging operators are derived from our merging framework. Furthermore, we propose a set of postulates for merging PLPs, some of which extend the postulates for merging classical knowledge bases, whilst others are specific to the merging of probabilistic beliefs.