27 pawns ready for action. A multi-indicator methodology and evaluation of thesaurus management tools from a LOD perspective
| DOI | https://doi.org/10.1108/LHT-11-2016-0123 |
| Date | 20 March 2017 |
| Pages | 99-119 |
| Published date | 20 March 2017 |
| Author | Gonzalo Mochón,Eva M. Méndez,Gema Bueno de la Fuente |
| Subject Matter | Library & information science,Librarianship/library management,Library technology,Information behaviour & retrieval,Information user studies,Metadata,Information & knowledge management,Information & communications technology,Internet |
27 pawns ready for action
A multi-indicator methodology and
evaluation of thesaurus management tools
from a LOD perspective
Gonzalo Mochón, Eva M. Méndez and Gema Bueno de la Fuente
Library and Information Science Department,
Universidad Carlos III de Madrid, Getafe, Spain
Abstract
Purpose –The purpose of this paper is to propose a methodology for assessing thesauri and other controlled
vocabularies management tools that can represent content using the Simple Knowledge Organization System
(SKOS) data model, and their use in a Linked Open Data (LOD) paradigm. It effectively analyses selected set
of tools in order to prove the validity of the method.
Design/methodology/approach –A set of 27 criteria grouped in five evaluation indicators is proposed and
applied to ten vocabulary management applications which are compliant with the SKOS data model. Previous
studies of controlled vocabulary management software are gathered and analyzed, to compare the evaluation
parameters used and the results obtained for each tool.
Findings –The results indicate t hat the tool that obtain s the highest score in ev ery indicator is
Poolparty. The second a nd third tools are, resp ectively, TemaTres an d Intelligent Theme Ma nager,
but scoring lower in most of t he evaluation items. The use of a broa d set of criteria to evaluate vocabula ries
management tools give s satisfactory result s. The set of five indica tors and 27 criteria pro posed
here represents a usefu l evaluation system in th e selection of current an d future tools to manage
vocabularies.
Research limitations/implications –The paper only assesses the ten most important/well know software
tools applied for thesaurus and vocabulary management until October 2016. However, the evaluation criteria
could be applied to new software that could appear in the future to create/manage SKOS vocabularies in
compliance with LOD standards.
Originality/value –The originality of this paper relies on the proposed indicators and criteria to evaluate
vocabulary management tools. Those criteria and indicators can be valuable also for future software that
might appear. The indicators are also applied to the most exhaustive and qualified list of this kind of tools.
The paper will help designers, information architects, metadata librarians, and other staff involved in the
design of digital information systems, to choose the right tool to manage their vocabularies in a LOD/
vocabulary scenario.
Keywords Linked open data, Software evaluation, Vocabularies, Semantic Web,
SKOS/RDF, Thesaurus management tools
Paper type Research paper
1. Introduction
Among the existing knowledge organization systems (KOS) we can find thesauri.
This kind of KOS can be defined as structured controlled vocabularies, whether generic
or for a specific theme, formed of concepts represented by terms, to which
meanings are given establishing equivalence among synonyms, and which are related
explicitly as described in ISO 25964-1:2011 (International Organization for
Standardization, 2011a).
Adapting thesauri to the web entailed a simplification in their management such as the
inclusion of search in interfaces and the transformation of their terms into hyper textual
links. The adoption of hypertext forms for thesaurus terms has been for some time the
main, and almost only, advantage of editing such vocabularies online (Hudon, 2003).
Nowadays many thesauri on the web show their content in a similar manner to
printed thesauri.
Library Hi Tech
Vol. 35 No. 1, 2017
pp. 99-119
© Emerald PublishingLimited
0737-8831
DOI 10.1108/LHT-11-2016-0123
Received 4 November 2016
Accepted 29 November 2016
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/0737-8831.htm
99
27 pawns
ready for
action
For more than 20 years, construction of web thesauri has taken place with various tools,
and for around 15 years the web has been practically the only channel of presentation for
these controlled vocabularies bringing these advantages:
•reduced cost: computerization of processes reduces the time needed for construction
and maintenance, leading to financial savings (Arano, 2005);
•usability testing helps understand users’comprehension of their function, as participants
in the creation, management, and optimization of thesauri (Arano and Codina, 2004;
Arano, 2005; Greenberg, 2004);
•improved access to terminological tools for helping with information retrieval
(Arano and Codina, 2004);
•the possibility to reuse terms for new projects or transfer their content to other
storage forms (Arano, 2005; Pastor-Sánchez et al., 2009); and
•utility in databases where each descriptor automatically links among associated
registries.
Despitetheir advantages, thesauri on theweb have limitations. One is relatedto the advantage
of transformingterms into hypertext links. Itwas an important development butnowadays it
is insufficient.Today many thesauri stillpresent their content as plaintext. Another limitation
is the use of website or database search by thesauri. These environments offer precision and
exhaustiveness that better matches user needs. García-Marco (2008) shows thesauri in the
internet must overcome localism and embed themselves in networks. One approach could be
using the Simple Knowledge Organization System (SKOS) data model.
SKOS is an OWL fullontology based on the Resource Description Framework(RDF). This
data model allows representations of organized knowledge, such as thesauri, to b e machine
readable,and interchanged by softwareapplications, thusnot just published “on”the web, but
being “of”the web (Méndez and Greenberg, 2012), thus increasing interoperability. SKOS
is a basic standard, simpler than OWL ontologies, and m ore economical (Pastor-Sánchez, 2011).
The SKOS data model is a World Wide Web Consortium (W3C) recommendation for
sharing and linking systematized knowledge on the web. Many KOS, including thesauri,
taxonomies, and classification systems have similar structures and are useful for similar
applications, facilitating interoperability.
SKOS offers a solution to the “localism”noted earlier by mapping concepts among
different vocabularies. Mapping vocabularies according to the requirements for linked data
will give greater possibilities to thesauri for knowledge discovery in the environments
where they are implemented.
The creation of data models shared through SKOS allows vocabulary users to search and
process archives in a uniform manner, but it does not solve the problems arising from the
design of applications for doing so (Tuominen et al., 2009).
Applications for constructing and managing controlled vocabularies that use SKOS
show a lack of uniformity in their functionality, as noted by Tuominen et al. (2009). Various
projects to migrate thesauri to SKOS use interfaces or tools created ad hoc such as:
•conversion of thesauri Integrated Public Sector Vocabulary, Medical Subject
Headings, and Gemeenschappelijke Thesaurus Audiovisuale Archieven, by
Van Assem et al. (2006);
•conversion of the Library of Congress Subject Headings in MARCXML to SKOS,
by Summers et al. (2008);
•representation of the Thesaurus for Social Sciences (TheSoz) by Zapilko and
Sure (2009);
100
LHT
35,1
To continue reading
Request your trial