NAISC-L: A Linked Data Interlinking Framework for Libraries, Archives and Museums
Citation:
MCKENNA, LUCY MARY, NAISC-L: A Linked Data Interlinking Framework for Libraries, Archives and Museums, Trinity College Dublin.School of Computer Science & Statistics, 2020Download Item:
Abstract:
This thesis presents a framework for Novel Authoritative Interlinking for Semantic Web Cataloguing in Libraries - or NAISC-L (pronounced noshk-el).
The Semantic Web (SW) is an extension of the current Web where data is given well defined meaning and where the relationships between data are defined in a common machine-readable format (Berners-Lee, Hendler, & Lassila, 2001). Linked Data (LD) describes a set of principles for publishing and engaging with data on the Semantic Web (Berners-Lee, 2006). A LD dataset is structured information encoded using the Resource Description Framework in which resources are identified using HTTP URIs. A LD dataset must contain links to related data (Berners-Lee, 2006) with the view of enhancing the knowledge associated with a specific entity (Papaleo, Pernelle, Sais & Dumont, 2014).
Within the Library, Archive and Museum (LAM) domain, LD interlinking could be used to enrich data and to improve data discoverability. However, upon reviewing of some of the leading LD projects in LAMs, there was a notable lack of interlinks created for purposes beyond authority control. Interlinking could also be used by LAMs to provide additional information and context for a resource. Literature suggests this lack of interlinking is, in part, due to the technical complexity of available LD tooling. There is also a lack of interlinking frameworks that support the creation of relationship links i.e. links to related resources that are not identical to the entity being interlinked.
Using the results of a LD questionnaire distributed to LAMs, a set of requirements for the development of a LD interlinking framework was defined. The proposed framework, NAISC-L, is comprised of an interlinking process and an accompanying tool. The interlinking process is cyclical in nature and consists of four stages - Entity Selection, Link-Type Selection, Provenance Data Generation, and RDF Graph Generation and Visualisation. The tool consists of a graphical user interface, a knowledge organisation approach which provides a structure for the linking of interlink and provenance data, and a provenance data model.
NAISC-L was evaluated, in terms of its effectiveness, efficiency and user satisfaction, via three usability tests - a think-aloud test, an interlink creation test and a field test. The participants of these evaluations were Information Professionals, including librarians, archivists and metadata cataloguers. Across all experiments, participants achieved a high level of accuracy when interlinking entities, and usability and utility measures indicated positive scores for efficiency and user satisfaction. Overall, NAISC-L was shown to be an effective framework for facilitating the creation of LD interlinks in the LAM domain.
This research has yielded one major contribution, the design, development and evaluation of the NAISC-L interlink framework, and two minor contributions. The first minor contribution is the provision of a report on the current state of LD in the LAM domain, and the second is NaiscProv. NaiscProv is an extension of the PROV Ontology, the W3C recommended standard for describing provenance data, which was developed to provide provenance descriptions for LD interlinks.
Sponsor
Grant Number
Science Foundation Ireland (SFI)
Author's Homepage:
https://tcdlocalportal.tcd.ie/pls/EnterApex/f?p=800:71:0::::P71_USERNAME:MCKENNL3Description:
APPROVED
Author: MCKENNA, LUCY MARY
Sponsor:
Science Foundation Ireland (SFI)Advisor:
O'Sullivan, DeclanPublisher:
Trinity College Dublin. School of Computer Science & Statistics. Discipline of Computer ScienceType of material:
ThesisCollections
Availability:
Full text availableKeywords:
Semantic Web, Libraries, Linked Data, Museums, ArchivesMetadata
Show full item recordLicences: