Unsupervised terminological ontology learning based on hierarchical topic modeling

Xiaofeng Zhu; Diego Klabjan; Patrick N. Bless

doi:10.1109/IRI.2017.18

Unsupervised terminological ontology learning based on hierarchical topic modeling

Xiaofeng Zhu, Diego Klabjan, Patrick N. Bless

Industrial Engineering and Management Sciences

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

5 Scopus citations

Abstract

In this paper, we present hierarchical relation-based latent Dirichlet allocation (hrLDA), a data-driven hierarchical topic model for extracting terminological ontologies from a large number of heterogeneous documents. In contrast to traditional topic models, hrLDA relies on noun phrases instead of unigrams, considers syntax and document structures, and enriches topic hierarchies with topic relations. Through a series of experiments, we demonstrate the superiority of hrLDA over existing topic models, especially for building hierarchies. Furthermore, we illustrate the robustness of hrLDA in the settings of noisy data sets, which are likely to occur in many practical scenarios. Our ontology evaluation results show that ontologies extracted from hrLDA are very competitive with the ontologies created by domain experts.

Original language	English (US)
Title of host publication	Proceedings - 2017 IEEE International Conference on Information Reuse and Integration, IRI 2017
Editors	Latifur Khan, Balaji Palanisamy, Chengcui Zhang, Sahra Sedigh Sarvestani
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	32-41
Number of pages	10
ISBN (Electronic)	9781538615621
DOIs	https://doi.org/10.1109/IRI.2017.18
State	Published - Nov 8 2017
Event	18th IEEE International Conference on Information Reuse and Integration, IRI 2017 - San Diego, United States Duration: Aug 4 2017 → Aug 6 2017

Publication series

Name	Proceedings - 2017 IEEE International Conference on Information Reuse and Integration, IRI 2017
Volume	2017-January

Other

Other	18th IEEE International Conference on Information Reuse and Integration, IRI 2017
Country/Territory	United States
City	San Diego
Period	8/4/17 → 8/6/17

Funding

This work was supported in part by Intel Corporation, Semiconductor Research Corporation (SRC). We are obliged to Professor Goce Trajcevski from Northwestern University for his insightful suggestions and discussions. This work was partly conducted using the Protege resource, which is supported by grant GM10331601 from the National Institute of General Medical Sciences of the United States National Institutes of Health.

Keywords

Hierarchical topic modeling
Knowledge acquisition
Ontology learning
Terminological ontology

ASJC Scopus subject areas

Computer Networks and Communications
Information Systems and Management
Artificial Intelligence
Computer Science Applications
Information Systems

Access to Document

10.1109/IRI.2017.18

Cite this

Zhu, X., Klabjan, D., & Bless, P. N. (2017). Unsupervised terminological ontology learning based on hierarchical topic modeling. In L. Khan, B. Palanisamy, C. Zhang, & S. S. Sarvestani (Eds.), Proceedings - 2017 IEEE International Conference on Information Reuse and Integration, IRI 2017 (pp. 32-41). (Proceedings - 2017 IEEE International Conference on Information Reuse and Integration, IRI 2017; Vol. 2017-January). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IRI.2017.18

Zhu, Xiaofeng ; Klabjan, Diego ; Bless, Patrick N. / Unsupervised terminological ontology learning based on hierarchical topic modeling. Proceedings - 2017 IEEE International Conference on Information Reuse and Integration, IRI 2017. editor / Latifur Khan ; Balaji Palanisamy ; Chengcui Zhang ; Sahra Sedigh Sarvestani. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 32-41 (Proceedings - 2017 IEEE International Conference on Information Reuse and Integration, IRI 2017).

@inproceedings{91b2c1d63c864e3897fce79a476698f0,

title = "Unsupervised terminological ontology learning based on hierarchical topic modeling",

abstract = "In this paper, we present hierarchical relation-based latent Dirichlet allocation (hrLDA), a data-driven hierarchical topic model for extracting terminological ontologies from a large number of heterogeneous documents. In contrast to traditional topic models, hrLDA relies on noun phrases instead of unigrams, considers syntax and document structures, and enriches topic hierarchies with topic relations. Through a series of experiments, we demonstrate the superiority of hrLDA over existing topic models, especially for building hierarchies. Furthermore, we illustrate the robustness of hrLDA in the settings of noisy data sets, which are likely to occur in many practical scenarios. Our ontology evaluation results show that ontologies extracted from hrLDA are very competitive with the ontologies created by domain experts.",

keywords = "Hierarchical topic modeling, Knowledge acquisition, Ontology learning, Terminological ontology",

author = "Xiaofeng Zhu and Diego Klabjan and Bless, {Patrick N.}",

note = "Publisher Copyright: {\textcopyright} 2017 IEEE.; 18th IEEE International Conference on Information Reuse and Integration, IRI 2017 ; Conference date: 04-08-2017 Through 06-08-2017",

year = "2017",

month = nov,

day = "8",

doi = "10.1109/IRI.2017.18",

language = "English (US)",

series = "Proceedings - 2017 IEEE International Conference on Information Reuse and Integration, IRI 2017",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "32--41",

editor = "Latifur Khan and Balaji Palanisamy and Chengcui Zhang and Sarvestani, {Sahra Sedigh}",

booktitle = "Proceedings - 2017 IEEE International Conference on Information Reuse and Integration, IRI 2017",

address = "United States",

}

Zhu, X, Klabjan, D & Bless, PN 2017, Unsupervised terminological ontology learning based on hierarchical topic modeling. in L Khan, B Palanisamy, C Zhang & SS Sarvestani (eds), Proceedings - 2017 IEEE International Conference on Information Reuse and Integration, IRI 2017. Proceedings - 2017 IEEE International Conference on Information Reuse and Integration, IRI 2017, vol. 2017-January, Institute of Electrical and Electronics Engineers Inc., pp. 32-41, 18th IEEE International Conference on Information Reuse and Integration, IRI 2017, San Diego, United States, 8/4/17. https://doi.org/10.1109/IRI.2017.18

Unsupervised terminological ontology learning based on hierarchical topic modeling. / Zhu, Xiaofeng; Klabjan, Diego; Bless, Patrick N.
Proceedings - 2017 IEEE International Conference on Information Reuse and Integration, IRI 2017. ed. / Latifur Khan; Balaji Palanisamy; Chengcui Zhang; Sahra Sedigh Sarvestani. Institute of Electrical and Electronics Engineers Inc., 2017. p. 32-41 (Proceedings - 2017 IEEE International Conference on Information Reuse and Integration, IRI 2017; Vol. 2017-January).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Unsupervised terminological ontology learning based on hierarchical topic modeling

AU - Zhu, Xiaofeng

AU - Klabjan, Diego

AU - Bless, Patrick N.

PY - 2017/11/8

Y1 - 2017/11/8

N2 - In this paper, we present hierarchical relation-based latent Dirichlet allocation (hrLDA), a data-driven hierarchical topic model for extracting terminological ontologies from a large number of heterogeneous documents. In contrast to traditional topic models, hrLDA relies on noun phrases instead of unigrams, considers syntax and document structures, and enriches topic hierarchies with topic relations. Through a series of experiments, we demonstrate the superiority of hrLDA over existing topic models, especially for building hierarchies. Furthermore, we illustrate the robustness of hrLDA in the settings of noisy data sets, which are likely to occur in many practical scenarios. Our ontology evaluation results show that ontologies extracted from hrLDA are very competitive with the ontologies created by domain experts.

AB - In this paper, we present hierarchical relation-based latent Dirichlet allocation (hrLDA), a data-driven hierarchical topic model for extracting terminological ontologies from a large number of heterogeneous documents. In contrast to traditional topic models, hrLDA relies on noun phrases instead of unigrams, considers syntax and document structures, and enriches topic hierarchies with topic relations. Through a series of experiments, we demonstrate the superiority of hrLDA over existing topic models, especially for building hierarchies. Furthermore, we illustrate the robustness of hrLDA in the settings of noisy data sets, which are likely to occur in many practical scenarios. Our ontology evaluation results show that ontologies extracted from hrLDA are very competitive with the ontologies created by domain experts.

KW - Hierarchical topic modeling

KW - Knowledge acquisition

KW - Ontology learning

KW - Terminological ontology

UR - http://www.scopus.com/inward/record.url?scp=85044223135&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85044223135&partnerID=8YFLogxK

U2 - 10.1109/IRI.2017.18

DO - 10.1109/IRI.2017.18

M3 - Conference contribution

AN - SCOPUS:85044223135

T3 - Proceedings - 2017 IEEE International Conference on Information Reuse and Integration, IRI 2017

SP - 32

EP - 41

BT - Proceedings - 2017 IEEE International Conference on Information Reuse and Integration, IRI 2017

A2 - Khan, Latifur

A2 - Palanisamy, Balaji

A2 - Zhang, Chengcui

A2 - Sarvestani, Sahra Sedigh

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 18th IEEE International Conference on Information Reuse and Integration, IRI 2017

Y2 - 4 August 2017 through 6 August 2017

ER -

Zhu X, Klabjan D, Bless PN. Unsupervised terminological ontology learning based on hierarchical topic modeling. In Khan L, Palanisamy B, Zhang C, Sarvestani SS, editors, Proceedings - 2017 IEEE International Conference on Information Reuse and Integration, IRI 2017. Institute of Electrical and Electronics Engineers Inc. 2017. p. 32-41. (Proceedings - 2017 IEEE International Conference on Information Reuse and Integration, IRI 2017). doi: 10.1109/IRI.2017.18

Unsupervised terminological ontology learning based on hierarchical topic modeling

Abstract

Publication series

Other

Funding

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this