Discovering rules by meta-level abduction

Authors:
Katsumi Inoue;Koichi Furukawa;Ikuo Kobayashi;Hidetomo Nabeshima
Affiliations:
National Institute of Informatics, Tokyo, Japan;SFC Research Institute, Keio University, Fujisawa, Japan;SFC Research Institute, Keio University, Fujisawa, Japan;Division of Medicine and Engineering Science, University of Yamanashi, Yamanashi, Japan
Venue:
ILP'09 Proceedings of the 19th international conference on Inductive logic programming
Year:
2009

Citing 10
Cited 3

A logical framework for default reasoning

Artificial Intelligence
Linear resolution for consequence finding

Artificial Intelligence
Minimal Answer Computation and SOL

JELIA '02 Proceedings of the European Conference on Logics in Artificial Intelligence
Theory Completion Using Inverse Entailment

ILP '00 Proceedings of the 10th International Conference on Inductive Logic Programming
Induction as Consequence Finding

Machine Learning
Consequence finding and computing answers with defaults

Journal of Intelligent Information Systems
Application of abductive ILP to learning metabolic network inhibition from temporal data

Machine Learning
Evaluating abductive hypotheses using an EM algorithm on BDDs

IJCAI'09 Proceedings of the 21st international jont conference on Artifical intelligence
SOLAR: An automated deduction system for consequence finding

AI Communications - Practical Aspects of Automated Reasoning
A consequence finding approach for full clausal abduction

DS'07 Proceedings of the 10th international conference on Discovery science

Hypothesizing about causal networks with positive and negative effects by meta-level abduction

ILP'10 Proceedings of the 20th international conference on Inductive logic programming
Completing causal networks by meta-level abduction

Machine Learning
Meta-interpretive learning of higher-order dyadic datalog: predicate invention revisited

IJCAI'13 Proceedings of the Twenty-Third international joint conference on Artificial Intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper addresses discovery of unknown relations from incomplete network data by abduction. Given a network information such as causal relations and metabolic pathways, we want to infer missing links and nodes in the network to account for observations. To this end, we introduce a framework of meta-level abduction, which performs abduction in the meta level. This is implemented in SOLAR, an automated deduction system for consequence finding, using a firstorder representation for algebraic properties of causality and the full-clausal form of network information and constraints. Meta-level abduction by SOLAR is powerful enough to infer missing rules, missing facts, and unknown causes that involve predicate invention in the form of existentially quantified hypotheses. We also show an application of rule abduction to discover certain physical techniques and related integrity constraints within the subject area of Skill Science.