ArThUR: A Tool for Markov Logic Network

Axel Bodart, Keyvin Evrard, James Jerson Ortiz Vega, Pierre Yves Schobbens

Research output: Contribution in Book/Catalog/Report/Conference proceedingConference contribution

258 Downloads (Pure)

Abstract

Logical approaches-and ontologies in particular-offer a welladapted framework for representing knowledge present on the Semantic Web (SW). These ontologies are formulated in Web Ontology Language (OWL2), which are based on expressive Description Logics (DL). DL are a subset of First-Order Logic (FOL) that provides decidable reasoning. Based on DL, it is possible to rely on inference mechanisms to obtain new knowledge from axioms, rules and facts specified in the ontologies. However, these classical inference mechanisms do not deal with : uncertainty probabilities. Several works recently targeted those issues (i.e. Pronto, PR-OWL, BayesOWL, etc.), but none of them combines OWL2 with Markov Logic Networks (MLN) formalism. Several open source software packages for MLN are available (e.g. Alchemy, Tuffy, RockIt, etc.). In this paper, we present ArThUR, a Java framework for reasoning with probabilistic information in the SW. ArThUR incorporate three open source software packages for MLN, which is able to reason with uncertainty information, showing that it can be used in several real-world domains. We also show several experiments of our tool with different ontologies.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PublisherSpringer Verlag
Pages319-328
Number of pages10
Volume8842
ISBN (Print)9783662455494
Publication statusPublished - 2014
EventInternational Workshops: OTM Academy,OTM Industry Case Studies Program, C and TC, EI2N, INBAST, ISDE, META4eS, MSC, and OnToContent 2014 - Amantea, Italy
Duration: 27 Oct 201431 Oct 2014

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume8842
ISSN (Print)03029743
ISSN (Electronic)16113349

Conference

ConferenceInternational Workshops: OTM Academy,OTM Industry Case Studies Program, C and TC, EI2N, INBAST, ISDE, META4eS, MSC, and OnToContent 2014
CountryItaly
CityAmantea
Period27/10/1431/10/14

Fingerprint

Ontology
Logic
Description Logics
Open Source Software
Semantic Web
Software Package
Software packages
Reasoning
Uncertainty
First-order Logic
Axioms
Java
Subset
Experiment
Experiments
Framework
Knowledge
Open source software

Cite this

Bodart, A., Evrard, K., Ortiz Vega, J. J., & Schobbens, P. Y. (2014). ArThUR: A Tool for Markov Logic Network. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8842, pp. 319-328). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8842). Springer Verlag.
Bodart, Axel ; Evrard, Keyvin ; Ortiz Vega, James Jerson ; Schobbens, Pierre Yves. / ArThUR: A Tool for Markov Logic Network. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8842 Springer Verlag, 2014. pp. 319-328 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{5b64727fa76b450a9e3b76e91ea5cc92,
title = "ArThUR: A Tool for Markov Logic Network",
abstract = "Logical approaches-and ontologies in particular-offer a welladapted framework for representing knowledge present on the Semantic Web (SW). These ontologies are formulated in Web Ontology Language (OWL2), which are based on expressive Description Logics (DL). DL are a subset of First-Order Logic (FOL) that provides decidable reasoning. Based on DL, it is possible to rely on inference mechanisms to obtain new knowledge from axioms, rules and facts specified in the ontologies. However, these classical inference mechanisms do not deal with : uncertainty probabilities. Several works recently targeted those issues (i.e. Pronto, PR-OWL, BayesOWL, etc.), but none of them combines OWL2 with Markov Logic Networks (MLN) formalism. Several open source software packages for MLN are available (e.g. Alchemy, Tuffy, RockIt, etc.). In this paper, we present ArThUR, a Java framework for reasoning with probabilistic information in the SW. ArThUR incorporate three open source software packages for MLN, which is able to reason with uncertainty information, showing that it can be used in several real-world domains. We also show several experiments of our tool with different ontologies.",
author = "Axel Bodart and Keyvin Evrard and {Ortiz Vega}, {James Jerson} and Schobbens, {Pierre Yves}",
year = "2014",
language = "English",
isbn = "9783662455494",
volume = "8842",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "319--328",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
address = "Germany",

}

Bodart, A, Evrard, K, Ortiz Vega, JJ & Schobbens, PY 2014, ArThUR: A Tool for Markov Logic Network. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 8842, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 8842, Springer Verlag, pp. 319-328, International Workshops: OTM Academy,OTM Industry Case Studies Program, C and TC, EI2N, INBAST, ISDE, META4eS, MSC, and OnToContent 2014, Amantea, Italy, 27/10/14.

ArThUR: A Tool for Markov Logic Network. / Bodart, Axel; Evrard, Keyvin; Ortiz Vega, James Jerson; Schobbens, Pierre Yves.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8842 Springer Verlag, 2014. p. 319-328 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8842).

Research output: Contribution in Book/Catalog/Report/Conference proceedingConference contribution

TY - GEN

T1 - ArThUR: A Tool for Markov Logic Network

AU - Bodart, Axel

AU - Evrard, Keyvin

AU - Ortiz Vega, James Jerson

AU - Schobbens, Pierre Yves

PY - 2014

Y1 - 2014

N2 - Logical approaches-and ontologies in particular-offer a welladapted framework for representing knowledge present on the Semantic Web (SW). These ontologies are formulated in Web Ontology Language (OWL2), which are based on expressive Description Logics (DL). DL are a subset of First-Order Logic (FOL) that provides decidable reasoning. Based on DL, it is possible to rely on inference mechanisms to obtain new knowledge from axioms, rules and facts specified in the ontologies. However, these classical inference mechanisms do not deal with : uncertainty probabilities. Several works recently targeted those issues (i.e. Pronto, PR-OWL, BayesOWL, etc.), but none of them combines OWL2 with Markov Logic Networks (MLN) formalism. Several open source software packages for MLN are available (e.g. Alchemy, Tuffy, RockIt, etc.). In this paper, we present ArThUR, a Java framework for reasoning with probabilistic information in the SW. ArThUR incorporate three open source software packages for MLN, which is able to reason with uncertainty information, showing that it can be used in several real-world domains. We also show several experiments of our tool with different ontologies.

AB - Logical approaches-and ontologies in particular-offer a welladapted framework for representing knowledge present on the Semantic Web (SW). These ontologies are formulated in Web Ontology Language (OWL2), which are based on expressive Description Logics (DL). DL are a subset of First-Order Logic (FOL) that provides decidable reasoning. Based on DL, it is possible to rely on inference mechanisms to obtain new knowledge from axioms, rules and facts specified in the ontologies. However, these classical inference mechanisms do not deal with : uncertainty probabilities. Several works recently targeted those issues (i.e. Pronto, PR-OWL, BayesOWL, etc.), but none of them combines OWL2 with Markov Logic Networks (MLN) formalism. Several open source software packages for MLN are available (e.g. Alchemy, Tuffy, RockIt, etc.). In this paper, we present ArThUR, a Java framework for reasoning with probabilistic information in the SW. ArThUR incorporate three open source software packages for MLN, which is able to reason with uncertainty information, showing that it can be used in several real-world domains. We also show several experiments of our tool with different ontologies.

UR - http://www.scopus.com/inward/record.url?scp=84910107369&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9783662455494

VL - 8842

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 319

EP - 328

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

PB - Springer Verlag

ER -

Bodart A, Evrard K, Ortiz Vega JJ, Schobbens PY. ArThUR: A Tool for Markov Logic Network. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8842. Springer Verlag. 2014. p. 319-328. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).