2014Unpublished venueRequires access

A Knowledge Representation Framework Based on Epistemic Logic

Teodor C. Przymusiński

Open publisher page 0 citations

Abstract

We introduce a uniform non-monotonic framework for knowledge representation based on epistemic logic which is sufficiently general to encompass several non-monotonic formalisms, including circumscription, autoepistemic logic, various semantics proposed for logic programs and deductive databases (stable semantics, well-founded semantics and stationary semantics) as well as Gelfond's epistemic specifications. The existence of such a uniform framework allows us not only to provide simpler and perhaps more natural definitions of various formalisms but it also enables us to better understand mutual relationships existing between them. Keywords: Non-monotonic reasoning, semantics of logic programs, disjunctive logic programs and deductive databases. 1 Introduction In recent years, various approaches to non-monotonic reasoning and different semantics for normal and disjunctive logic programs have been proposed, including: ffl Autoepistemic Logic [Moo85]; ffl Circumscription, CWA, GCWA, EC...

About this research paper

What this paper is about

We introduce a uniform non-monotonic framework for knowledge representation based on epistemic logic which is sufficiently general to encompass several non-monotonic formalisms, including circumscription, autoepistemic logic, various semantics proposed for logic programs and deductive databases (stable semantics, well-founded semantics and stationary semantics) as well as Gelfond's epistemic specifications. The existence of such a uniform framework allows us not only to provide simpler and perhaps more natural definitions of various formalisms but it also enables us to better understand mutual relationships existing between them. Keywords: Non-monotonic reasoning, semantics of logic programs, disjunctive logic programs and deductive databases. 1 Introduction In recent years, various approaches to non-monotonic reasoning and different semantics for normal and disjunctive logic programs have been proposed, including: ffl Autoepistemic Logic [Moo85]; ffl Circumscription, CWA, GCWA, EC...

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

We introduce a uniform non-monotonic framework for knowledge representation based on epistemic logic which is sufficiently general to encompass several non-monotonic formalisms, including circumscription, autoepistemic logic, various semantics proposed for logic programs and deductive databases (stable semantics, well-founded semantics and stationary semantics) as well as Gelfond's epistemic specifications. The existence of such a uniform framework allows us not only to provide simpler and perhaps more natural definitions of various formalisms but it also enables us to better understand mutual relationships existing between them. Keywords: Non-monotonic reasoning, semantics of logic programs, disjunctive logic programs and deductive databases. 1 Introduction In recent years, various approaches to non-monotonic reasoning and different semantics for normal and disjunctive logic programs have been proposed, including: ffl Autoepistemic Logic [Moo85]; ffl Circumscription, CWA, GCWA, EC...

Key concepts: Rotation formalisms in three dimensions, Circumscription, Non-monotonic logic, Knowledge representation and reasoning, Autoepistemic logic, Stable model semantics, Computer science, Semantics (computer science)

Related papers

Back to paper searchBrowse research topicsOriginal source
A Knowledge Representation Framework Based on Epistemic Logic — Research Paper | ScholarLens