Semantics of interworking knowledge based systems

Georg Reichwein, José Fiadeiro

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Mixed institutions are proposed as a semantic framework for interworking heterogeneous knowledge based systems, where heterogeneity means that the components of a knowledge base may be based on different logical systems. Using deductive databases as an example, we illustrate the use of this framework for extending relational databases with inference capabilities in first order and temporal logic. Taking institutions as the formal notion of a logical system, a novel institution of relational algebra is introduced together with an institution morphism which formally relates it to first-order logic. Reasoning about dynamics is encompassed by a general construction of a temporal logic institution over an arbitrary base institution. The results are relevant for a declarative semantics of interoperable knowledge bases and multi-agent systems.

Original languageEnglish
Title of host publicationProgress in Artificial Intelligence - 6th Portuguese Conference on AI, EPIA 1993, Proceedings
EditorsMiguel Filgueiras, Luis Damas
Place of PublicationBerlin
PublisherSpringer Verlag
Pages93-104
Number of pages12
Volume727
ISBN (Electronic)9783540480365
ISBN (Print)9783540572879
DOIs
Publication statusPublished - 23 Sept 1993
Event6th Portuguese Conference on Artificial Intelligence, EPIA 1993 - Porto, Portugal
Duration: 6 Oct 19938 Oct 1993

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume727 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference6th Portuguese Conference on Artificial Intelligence, EPIA 1993
Country/TerritoryPortugal
CityPorto
Period6/10/938/10/93

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Semantics of interworking knowledge based systems'. Together they form a unique fingerprint.

Cite this