Please use this identifier to cite or link to this item:
Author(s): Câmara, A.
Almeida, A. de.
Oliveira, J.
Date: 2024
Title: Transforming the CIDOC-CRM model into a megalithic monument property graph
Journal title: Journal of Computer Applications in Archaeology
Volume: 7
Number: 1
Pages: 213 - 224
Reference: Câmara, A., Almeida, A. de., & Oliveira, J. (2024). Transforming the CIDOC-CRM model into a megalithic monument property graph. Journal of Computer Applications in Archaeology, 7(1), 213-224.
ISSN: 2514-8362
DOI (Digital Object Identifier): 10.5334/jcaa.151
Keywords: Knowledge graph
Labelled property graph
Abstract: This paper presents a method to store information about megalithic monument-building components as graph nodes in a knowledge graph (KG). As a case study, we analyse the dolmens from the region of Pavia (Portugal). To build the KG, information has been extracted from unstructured data to populate a schema model based on the International Committee for Documentation – Conceptual Reference Model (CIDOC-CRM). In order to prepare the archaeological monument’s information for bulk loading, it was transformed into semi-structured data. While the semi-structured file was used to populate the classes with their respective properties and instances, the KG labels and types were defined using some of the entities and relations defined by the CIDOC-CRM. The knowledge-driven model was built to represent dolmens in a formal and structured manner using Neo4j, a property-graph database. Modelling a labelled property graph based on predefined labels as a KG enables to transform textual semantic data into instances and properties. Thus, we show that it is possible to represent at a granular level all the information about the structural components of monuments since heterogeneities, granularities, and large amounts of data can be handled by a KG. Therefore, a KG implemented using a native graph database can improve data storage and processing, making it interoperable between humans, humans and machines and machine to machine.
Peerreviewed: yes
Access type: Open Access
Appears in Collections:ISTAR-RI - Artigos em revistas científicas internacionais com arbitragem científica

Files in This Item:
File SizeFormat 
article_103671.pdf1,49 MBAdobe PDFView/Open

FacebookTwitterDeliciousLinkedInDiggGoogle BookmarksMySpaceOrkut
Formato BibTex mendeley Endnote Logotipo do DeGóis Logotipo do Orcid 

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.