Building Lightweight Ontologies for Faceted Search with Named Entity Recognition: Case WarMemoirSampo

Mikko Koho; Rafael Leal; Esko Ikkala; Minna Tamper; Heikki Rantala; Eero Hyvönen

Building Lightweight Ontologies for Faceted Search with Named Entity Recognition: Case WarMemoirSampo

Mikko Koho^*, Rafael Leal, Esko Ikkala, Minna Tamper, Heikki Rantala, Eero Hyvönen

^*Corresponding author for this work

Research output: Contribution to journal › Conference article › Scientific › peer-review

1 Citation (Scopus)

24 Downloads (Pure)

Abstract

This paper discusses building lightweight ontologies for faceted search user interfaces with Named Entity Recognition (NER) from textual data. This is studied in the context of building a Knowledge Graph for the textual indexing of interview videos in the in-use WarMemoirSampo system, consisting of a Linked Open Data service and an open semantic web portal for contextualized video viewing. It is shown that state-of-the-art NER tools are able to find entities from textual data and categorize them with high enough recall and precision to be useful for building facet ontologies, without involving considerable manual domain ontology engineering. To enable entity disambiguation and to be able to show relevant contextual information and useful links for the users of the portal, also Named Entity Linking techniques are employed.

Original language	English
Pages (from-to)	19-35
Number of pages	17
Journal	CEUR Workshop Proceedings
Volume	3184
Publication status	Published - 11 Aug 2022
MoE publication type	A4 Article in a conference publication
Event	International Workshop on Knowledge Graph Generation from Text - Hersonissos, Greece Duration: 30 May 2022 → 30 May 2022 Conference number: 1

Keywords

Faceted Search
Information Extraction
Linked Data
Named Entity Linking
Named Entity Recognition
Ontologies

Access to Document

SCI_Koho_etal_TEXT2KGFinal published version, 1.31 MBLicence: CC BY

http://ceur-ws.org/Vol-3184/TEXT2KG_Paper_2.pdfLicence: CC BY

OpenUrl availability

Full text

Cite this

@article{ac3830667e244a84bbc3481c14383a83,

title = "Building Lightweight Ontologies for Faceted Search with Named Entity Recognition: Case WarMemoirSampo",

abstract = "This paper discusses building lightweight ontologies for faceted search user interfaces with Named Entity Recognition (NER) from textual data. This is studied in the context of building a Knowledge Graph for the textual indexing of interview videos in the in-use WarMemoirSampo system, consisting of a Linked Open Data service and an open semantic web portal for contextualized video viewing. It is shown that state-of-the-art NER tools are able to find entities from textual data and categorize them with high enough recall and precision to be useful for building facet ontologies, without involving considerable manual domain ontology engineering. To enable entity disambiguation and to be able to show relevant contextual information and useful links for the users of the portal, also Named Entity Linking techniques are employed.",

keywords = "Faceted Search, Information Extraction, Linked Data, Named Entity Linking, Named Entity Recognition, Ontologies",

author = "Mikko Koho and Rafael Leal and Esko Ikkala and Minna Tamper and Heikki Rantala and Eero Hyv{\"o}nen",

note = "Funding Information: Ilpo Murtovaara, Markus Merenmies, and Kare Salonvaara contributed in creating the videos and their metadata used in the work of this paper. Tammenlehv{\"a}n Perinneliitto ry with the National Archives of Finland funded our work. The work is partly related and funded by the EU project InTaVia: In/Tangible European Heritage20, and the EU COST action Nexus Linguarum21 on linguistic data science. The authors acknowledge CSC – IT Center for Science, Finland, for providing for computational resources. Publisher Copyright: {\textcopyright} 2022 Copyright for this paper by its authors. | openaire: EC/H2020/101004825/EU//InTaVia; International Workshop on Knowledge Graph Generation from Text, Text2KG ; Conference date: 30-05-2022 Through 30-05-2022",

year = "2022",

month = aug,

day = "11",

language = "English",

volume = "3184",

pages = "19--35",

journal = "CEUR WORKSHOP PROCEEDINGS",

issn = "1613-0073",

publisher = "RWTH Aachen University",

}

TY - JOUR

T1 - Building Lightweight Ontologies for Faceted Search with Named Entity Recognition

T2 - International Workshop on Knowledge Graph Generation from Text

AU - Koho, Mikko

AU - Leal, Rafael

AU - Ikkala, Esko

AU - Tamper, Minna

AU - Rantala, Heikki

AU - Hyvönen, Eero

N1 - Conference code: 1

PY - 2022/8/11

Y1 - 2022/8/11

N2 - This paper discusses building lightweight ontologies for faceted search user interfaces with Named Entity Recognition (NER) from textual data. This is studied in the context of building a Knowledge Graph for the textual indexing of interview videos in the in-use WarMemoirSampo system, consisting of a Linked Open Data service and an open semantic web portal for contextualized video viewing. It is shown that state-of-the-art NER tools are able to find entities from textual data and categorize them with high enough recall and precision to be useful for building facet ontologies, without involving considerable manual domain ontology engineering. To enable entity disambiguation and to be able to show relevant contextual information and useful links for the users of the portal, also Named Entity Linking techniques are employed.

AB - This paper discusses building lightweight ontologies for faceted search user interfaces with Named Entity Recognition (NER) from textual data. This is studied in the context of building a Knowledge Graph for the textual indexing of interview videos in the in-use WarMemoirSampo system, consisting of a Linked Open Data service and an open semantic web portal for contextualized video viewing. It is shown that state-of-the-art NER tools are able to find entities from textual data and categorize them with high enough recall and precision to be useful for building facet ontologies, without involving considerable manual domain ontology engineering. To enable entity disambiguation and to be able to show relevant contextual information and useful links for the users of the portal, also Named Entity Linking techniques are employed.

KW - Faceted Search

KW - Information Extraction

KW - Linked Data

KW - Named Entity Linking

KW - Named Entity Recognition

KW - Ontologies

UR - http://www.scopus.com/inward/record.url?scp=85137149182&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:85137149182

VL - 3184

SP - 19

EP - 35

JO - CEUR WORKSHOP PROCEEDINGS

JF - CEUR WORKSHOP PROCEEDINGS

SN - 1613-0073

Y2 - 30 May 2022 through 30 May 2022

ER -

Building Lightweight Ontologies for Faceted Search with Named Entity Recognition: Case WarMemoirSampo

Abstract

Keywords

Access to Document

OpenUrl availability

Other files and links

Fingerprint

InTaVia: In/Tangible European Heritage – Visual Analysis, Curation and Communication

Cite this

Building Lightweight Ontologies for Faceted Search with Named Entity Recognition: Case WarMemoirSampo

Abstract

Keywords

Access to Document

OpenUrl availability

Other files and links

Fingerprint

Projects

InTaVia: In/Tangible European Heritage – Visual Analysis, Curation and Communication

Cite this