TBox learning from incomplete data by inference in BelNet+

Man Zhu*, Zhiqiang Gao, Jeff Z. Pan, Yuting Zhao, Ying Xu, Zhibin Quan

*Corresponding author for this work

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

In this work we deal with the problem of TBox learning from incomplete semantic web data. TBox, or conceptual schema, is the backbone of a Description Logic (DL) ontology, but is always difficult to obtain. Existing approaches either fail in getting correct results under incompleteness or learn results that are not enough to resolve the incompleteness. We propose to transform TBox learning in DL into inference in the extension of Bayesian Description Logic Network (abbreviated as BelNet+), whereby the structure in the data is leveraged when evaluating the relationships between two concepts. BelNet+, integrating the probabilistic inference capability of Bayesian Networks with the logical formalism of DL ontologies - Description Logics, supports promising inference. In this paper, we firstly explain the details of BelNet+ and introduce a TBox learning approach based on BelNet+. In order to overcome the drawbacks of current evaluation metrics, we then propose a novel evaluation framework conforming to the Open World Assumption (OWA) generally made in the semantic web. Finally the results from empirical studies on comparisons with the state-of-the-art TBox learners verify the effectiveness of our approach.

Original languageEnglish
Pages (from-to)30-40
Number of pages11
JournalKnowledge-Based Systems
Volume75
Issue numberC
Early online date18 Nov 2014
DOIs
Publication statusPublished - 1 Feb 2015

Fingerprint

Semantic Web
Ontology
Bayesian networks
Incomplete data
Inference
Description logics
Semantic web
Incompleteness
Evaluation

Keywords

  • Evaluation framework
  • Ontology learning
  • Probabilistic description logics
  • Semantic web
  • TBox learning

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence
  • Management Information Systems
  • Information Systems and Management

Cite this

TBox learning from incomplete data by inference in BelNet+. / Zhu, Man; Gao, Zhiqiang; Pan, Jeff Z.; Zhao, Yuting; Xu, Ying; Quan, Zhibin.

In: Knowledge-Based Systems, Vol. 75, No. C, 01.02.2015, p. 30-40.

Research output: Contribution to journalArticle

Zhu, Man ; Gao, Zhiqiang ; Pan, Jeff Z. ; Zhao, Yuting ; Xu, Ying ; Quan, Zhibin. / TBox learning from incomplete data by inference in BelNet+. In: Knowledge-Based Systems. 2015 ; Vol. 75, No. C. pp. 30-40.
@article{9a6428a33f9d48fca3d1e4269e8b15a2,
title = "TBox learning from incomplete data by inference in BelNet+",
abstract = "In this work we deal with the problem of TBox learning from incomplete semantic web data. TBox, or conceptual schema, is the backbone of a Description Logic (DL) ontology, but is always difficult to obtain. Existing approaches either fail in getting correct results under incompleteness or learn results that are not enough to resolve the incompleteness. We propose to transform TBox learning in DL into inference in the extension of Bayesian Description Logic Network (abbreviated as BelNet+), whereby the structure in the data is leveraged when evaluating the relationships between two concepts. BelNet+, integrating the probabilistic inference capability of Bayesian Networks with the logical formalism of DL ontologies - Description Logics, supports promising inference. In this paper, we firstly explain the details of BelNet+ and introduce a TBox learning approach based on BelNet+. In order to overcome the drawbacks of current evaluation metrics, we then propose a novel evaluation framework conforming to the Open World Assumption (OWA) generally made in the semantic web. Finally the results from empirical studies on comparisons with the state-of-the-art TBox learners verify the effectiveness of our approach.",
keywords = "Evaluation framework, Ontology learning, Probabilistic description logics, Semantic web, TBox learning",
author = "Man Zhu and Zhiqiang Gao and Pan, {Jeff Z.} and Yuting Zhao and Ying Xu and Zhibin Quan",
note = "Acknowledgements This work is partially funded by the National Science Foundation of China under Grant 61170165, the EU IAPP K-Drive project (286348) and the EPSRC WhatIf project (EP/J014354/1). The authors would like to thank Campbell Wilson for proof-reading the document.",
year = "2015",
month = "2",
day = "1",
doi = "10.1016/j.knosys.2014.11.004",
language = "English",
volume = "75",
pages = "30--40",
journal = "Knowledge-Based Systems",
issn = "0950-7051",
publisher = "Elsevier",
number = "C",

}

TY - JOUR

T1 - TBox learning from incomplete data by inference in BelNet+

AU - Zhu, Man

AU - Gao, Zhiqiang

AU - Pan, Jeff Z.

AU - Zhao, Yuting

AU - Xu, Ying

AU - Quan, Zhibin

N1 - Acknowledgements This work is partially funded by the National Science Foundation of China under Grant 61170165, the EU IAPP K-Drive project (286348) and the EPSRC WhatIf project (EP/J014354/1). The authors would like to thank Campbell Wilson for proof-reading the document.

PY - 2015/2/1

Y1 - 2015/2/1

N2 - In this work we deal with the problem of TBox learning from incomplete semantic web data. TBox, or conceptual schema, is the backbone of a Description Logic (DL) ontology, but is always difficult to obtain. Existing approaches either fail in getting correct results under incompleteness or learn results that are not enough to resolve the incompleteness. We propose to transform TBox learning in DL into inference in the extension of Bayesian Description Logic Network (abbreviated as BelNet+), whereby the structure in the data is leveraged when evaluating the relationships between two concepts. BelNet+, integrating the probabilistic inference capability of Bayesian Networks with the logical formalism of DL ontologies - Description Logics, supports promising inference. In this paper, we firstly explain the details of BelNet+ and introduce a TBox learning approach based on BelNet+. In order to overcome the drawbacks of current evaluation metrics, we then propose a novel evaluation framework conforming to the Open World Assumption (OWA) generally made in the semantic web. Finally the results from empirical studies on comparisons with the state-of-the-art TBox learners verify the effectiveness of our approach.

AB - In this work we deal with the problem of TBox learning from incomplete semantic web data. TBox, or conceptual schema, is the backbone of a Description Logic (DL) ontology, but is always difficult to obtain. Existing approaches either fail in getting correct results under incompleteness or learn results that are not enough to resolve the incompleteness. We propose to transform TBox learning in DL into inference in the extension of Bayesian Description Logic Network (abbreviated as BelNet+), whereby the structure in the data is leveraged when evaluating the relationships between two concepts. BelNet+, integrating the probabilistic inference capability of Bayesian Networks with the logical formalism of DL ontologies - Description Logics, supports promising inference. In this paper, we firstly explain the details of BelNet+ and introduce a TBox learning approach based on BelNet+. In order to overcome the drawbacks of current evaluation metrics, we then propose a novel evaluation framework conforming to the Open World Assumption (OWA) generally made in the semantic web. Finally the results from empirical studies on comparisons with the state-of-the-art TBox learners verify the effectiveness of our approach.

KW - Evaluation framework

KW - Ontology learning

KW - Probabilistic description logics

KW - Semantic web

KW - TBox learning

UR - http://www.scopus.com/inward/record.url?scp=84920538184&partnerID=8YFLogxK

U2 - 10.1016/j.knosys.2014.11.004

DO - 10.1016/j.knosys.2014.11.004

M3 - Article

AN - SCOPUS:84920538184

VL - 75

SP - 30

EP - 40

JO - Knowledge-Based Systems

JF - Knowledge-Based Systems

SN - 0950-7051

IS - C

ER -