 |
Maysah Alsahar, Bassel AlKhatib Syrian Virtual University
Syrian Arab Republic
{maysah_alsahar@hotmail.com} {t_balkhatib@svuonline.org}
ABSTRACT: The use of semantic web technologies is very limited in Arab libraries, even with the existence of important historical and literary documents, and the great digitization movement now, the documents are still trapped within the databases of the separated libraries and are not visible to search engines. The research aims to Study the applications of semantic web technologies in catalogs of digital, hybrid libraries and catalogs of public libraries, exploring the main ontologies and conceptual Data models used in the field, and the possibilities of reusing them with Arab libraries, or the need to build new ontologies, studying the necessary steps, tools, and best practices, define the difficulties and challenges facing the Arabic semantic library and searching for possible technical solutions, our experiments target the heterogeneous databases of small, medium or large libraries, considering that some Arabic library catalogs are not following a standard Data Model, to contribute to the research and development of integrating bibliographic data of Arab libraries and moving towards the Semantic web and Linked Open Data.
Subject Categories and Descriptors: [H.3.7 Digital Libraries] Standards [H.3.3 Information Search and Retrieval
General Terms: Digital Libraries, Semantic Web, Arabic Digital Libraries, RDF
Keywords: Semantic web, RDB to RDF, Cataloging, library Database, BIBFRAME
Received: 19 September 2022, Revised 21 December 2022, Ac-
cepted 2 March 2023
Review Metrics: Review Scale: 0/6, Review Score 5.04, Inter-reviewer Consistency: 84.6%
DOI: 10.6025/jdim/2023/21/2/29-38
Library management systems are huge and complex systems that include many functions such as membership management, billing, loan operations, catalogue management, etc., there are a lot of open source Integrated Library Systems that can be customized to serve our choices towards Linked Open Data, some of the well-
known examples are Koha, Dspace, and Fedora1, recent researches in the field of libraries deal with many concepts such as NLP2, NER/NEE3 cross-lingual ontology enrichment techniques, Semantic search, Recommendation systems, Ontology-based automatic text classification.. etc., but cataloging remains the heart and foundation of Library work, and traditional databases with
1https://fedora.lyrasis.org/ 2Natural Language processing
3Named entity Recognition/Named Entity Extraction
traditional search based on Considering names as literal strings are insufficient now, catalogs of most of the libraries in the Arabic region are still a part of the Deep Web, so the need for Semantic Web technologies in libraries has become urgent, In addition, The sources of information between libraries are similar, and efforts are repeated in preparing the same sources of information, which leads to duplication of effort and loss of time [1], Because of the different methods of modeling data, the library community has developed many standards to unify the methods to deal with physical and digital containers, such as (AACR24, RDA5, MARC6, FRBR7, BIBFRAME, etc.), and
all new standards and data models adopt the principles of Linked Data.
Means sharing data and facts rather than sharing the text of a page. It was first given by Sir Tim Berners-Lee in 2001. And it helps build a technology stack to support a ‘web of data’ rather than a ‘web of documents’[2], the basic data model in the semantic web is the Resource Description Framework (RDF) and SPARQL is the W3C recommendation query language for RDF graphs.
According to the IFLA/UNESCO Manifesto for Digital Libraries [3], “a digital library is an online collection of digital objects, of assured quality, that are created or collected and managed according to internationally accepted principles for collection development and made accessible in a coherent and sustainable manner, supported by services necessary to allow users to retrieve and exploit the resources”, and being ‘semantic’ means that all data stored in the library is available in RDF for querying and processing by other applications[4]. In fact, this definition is too idealistic and our current goal is to have library catalogs at least stored in RDF format with unified URIs for Names and main concepts, with consideration to the relationships between different intellectual works such as reprint, translation, review, criticism, and the relations between other main catalog entities like persons, organizations, events, etc.
A word consisting of two syllables, ‘biblion’, means book, and ‘graphía’, means writing, the simplest definition of the word is a description of books, Bibliography includes the name of the author, the title of the information container, the edition, the publication data, the number of
pages, and the type of the information container (a book, a document, a map, etc.).
They describe metadata standards, such as Bibliographic ontology BIBO and MarcOnt, according to [8] Main LOD vocabularies used by libraries to publish bibliographic information are: BIBFRAME, BIBO, FRBRoo, Dublin Core, EDM, FRBR, LRM, RDA, and Schema.org.
They are useful as knowledge sources that define the meaning of most domain concepts, their hierarchy, properties, and relationships, for instance, OntoMath [9]is related to the development of specialized ontologies in the field of mathematics, while we find in [7] a sample of creating domain ontologies for Bulgarian folk songs, orga-
nized in a digital library called DjDL, LCSH8 also is a famous example.
are oriented to the description of the different types of users, their requirements, and interactions, and the most well-known ontology here is foaf.
Is a bibliographic data model, published by the Library of Congress on 2012, and considered as the future of libraries [10]and the successor of MARC10 which has been the dominant library data model for the last five decades, BIBFRAME is an ontological data model for bibliographic description, based on the principles of linked data and
semantic web technologies; it supports the interoperability
between libraries Network that requires a high degree of data analysis and relationships between them; and improves the quality of the information retrieval process across the web environment, the higher levels of abstraction of the bibliographic data representation model according to BIBFRAME 2.0 are shown in Fig.1 as: Work class and is used to express the same intellectual work that the indexed information source includes, Instance class is used to express the embodiment or the material of the intellectual work, Item or indexed item (for example, The Merchant of Venice is an intellectual work, and the novel or film represents it is Instance, while the copy owned by the library is the Item).
4Anglo-American Cataloguing Rules 5Resource Description and Access 6Machine-Readable Cataloging
7Functional Requirements for Bibliographic Records 8Library of Congress Subject Headings 9https://www.loc.gov/bibframe/docs/bibframe2-model.html 10Machine Readable Catalogue
Figure 1. BIBFRAME2 higher level of abstraction
Integrating heterogeneous databases is a challenge for organizations and researchers[11] and in the field of libraries the solution to this problem was _for a long time_ using the Union Catalog [1] where a group of libraries can translate their records into a shared format _which was MARC in most cases_ so they can communicate with each other, which is similar to ETL (Extract, transform, load) the idea is still the same, but we use ontology as a shared understanding between databases of different libraries, so mapping between databases and ontology generate RDF triples that can be queried by SPARQL, Two ways of data access and integration are usually used: the first is materialization: where data are transformed taking into account the mappings and the ontologies(for example, data are converted into RDF and loaded into a triple store, so that it can be natively queried using SPARQL) just like the usual ETL(Extract, Transform, Load), and the second way is virtualization: where the transformation is done on the queries using the mapping rules and ontologies, which can then be evaluated on the original data sources.[12].
W3C recommended using Direct mapping where each table maps to a class, each row maps to an instance, and each column maps to a data type property, this way is quick and easy but causes a lot of redundancy, the
other recommendation is using R2RML11 or RML and now there is some easier languages like YARRRML [13], in
11https://www.w3.org/TR/r2rml/
addition to a diversity of converters to RDF or Mapping tools like Openrefine, Ontop plugin+ Protégé, Morph family (Morph-CSV, Morph-RDB, Mapeathor..)[12], D2RQ[14], RDB2RDF plugin for Eclipse.
The research required a review of the most important data models and ontologies in the field, discovering tools and techniques, and understanding the potentials and challenges in Arab libraries to reach the semantic Library.
Many researchers have discussed the importance of the Semantic Web and its benefits related to digital libraries in a theoretical way without going into the implementation mechanisms and practical application (i.e. review of literature, surveys) such[10], [15]–[19], while others were more focused on Bibliographic ontologies [1], [14], [20]–[26], most of the recent researches in this field study Mapping between one Data Model with another, in particular, MARC to BIBFRAME [26]–[28], or between EDM and BIBFRAME [23], [29], [30], these studies focus on the transition from standard data models to BIBFRAME, while the current research aims to work on small, medium, or even large libraries that use some old databases that do not adopt any known standard, depending on mapping mechanisms between databases and ontology to obtain the knowledge graph [12]. many Artificial Intelligence researches studied the document automatic classification based on measuring the similarity [31], and some Arabic researchers combine natural language processing of the Arabic language with automatic classification and automatic ontology building [32], [33] which is more
Paper | Pub Date | Standard Data Models/ reused ontologies | Methodology/ main Programing tools | Main Ontology type (Library perspective) |
[15] | 2011 | Principles and guidelines | ||
[1] | 2013 | FRBR, MARC | Java + jena library, Protégé | bibliographic |
[33] | 2014 | Java + jena library, Protégé | subject | |
[29] | 2014 | EDM, BIBFRAME | Mapping ontologies | Bibliographic, subject |
[37] | 2014 | AGROVOC, CAB Thesaurus, ASFA Thesaurus | ||
[7] | 2015 | ASP.NET MVC5, Protégé | subject | |
[26] | 2015 | Dublin Core, BIBFRAME | Open refine +RDF extension, Apache Jena | Bibliographic |
[10] | 2016 | BIBFRAME, MARC | Bibliographic | |
[20] | 2017 | OAI-PMH VIVO, BIBO, FOAF, SKOS, geonames, Dublin Core, Vitro, and Marc 21. | ABCD, Moodle and DSpace VIVO software, PHP, MySQL | Bibliographic |
[27] | 2017 | BIBFRAME, RDA, Schema.org, CIDOC CRM. | LD4P Labs | Bibliographic |
[38] | 2017 | BIBFRAME, EDM, FRBR, FRBRoo | Review of literature | Bibliographic |
[39] | 2017 | FaBiO, CiTO, BiRO, C4O DoCO, PRO, PSO, PWO | PHP-script, VBA macro | Bibliographic |
[17] | 2017 | Review of literature | ||
[40] | 2018 | Interviews | ||
[14] | 2018 | FABIO, SWRC, Dublin Core | D2RQ Pubby linked data interface Grails framework | Bibliographic, community-aware ontology |
[41] | 2019 | OAI-PMH | R2RML, Jena Fuseki triplestore,, ElasticSearch server, Pubby tool | Bibliographic |
[18] | 2018 | Review of literature | ||
[36] | 2019 | MARC | ||
[21] | 2019 | CIDOC CRM, FRBR, BIBO, HiCO, CiTO, AAT, FaBiO,Biro, BCO | Bibliographic | |
[19] | 2019 | Review of literature | Bibliographic | |
[30] | 2020 | BIBFRAME2.0, RDA | Python, XSLT, Protégé, Virtuoso RDF server | Bibliographic |
[42] | 2020 | (LRM12) + aspects of (FRBRoo), DOing REusable MUSic (DOREMUS) | Protégé SPARQL | subject |
[22] | 2020 | EDM, LIDO | Europeana portal, eCHO Framework, JAVA | bibliographic |
[11] | 2020 | Jena API, Protégé | ||
[32] | 2020 | GATE, Arabic WordNet (AWN), Protégé | subject |
[28] | 2020 | MARC, Schema.org, BIBFRAME | HTRC’s suite of tools, (VIAF13) through its API, LoC’s subject heading API | bibliographic |
[24] | 2020 | SHACL14 | bibliographic | |
[43] | 2020 | Similarity measures, Jena Subject (domain ontology) | ||
[31] | 2021 | Document Classification | ||
[12] | 2021 | Mapping Languages, Morph_RDB, Ontop. | ||
[44] | 2022 | OpenRefine | bibliographic |
Table 1. A Sample of the Related Works with the Main Reused Data Models and Methodologies.
useful in the side of subject ontologies rather than bibliographic ontologies where we are depending on some strong, well-defined ontologies that built by the experts of the field, and many Arabic researchers assure the urgent need for developing Arabic Ontologies to be exploited in the Arabic SW applications [34]. one of the serious attempts in the Arab region to assure the importance of directing libraries toward LOD is the conference titled “The Standard Model "BIBFRAME" for Resources Description and Access in Web Environment” [35] (proceedings only available in Arabic language ), the work on [36] discussed interoperability of data and mapping between ‘Winisis’ system and Koha system in the library of Damascus university, but to the best of our knowledge, there is no Arabic research that tried to exploit BIBFRAME ontology to achieve the semantic interoperability between libraries,(i.e.
Theory has not been put into practice). Most Arabic researches were written from the point of view of library specialists and focused on conceptual aspects such as [10], and did not go into the practical steps of the transition to linked data, except for some rare cases like [1], [32], [33] where the authors built their own ontologies without exploiting the benefits of reusing standard data models. Table1 shows a sample of the reviewed papers to view the most popular Data Models and tools in the field.
We presume having multiple heterogenous library databases, and their data will go through the steps as shown in figure 2.
Figure 2. Main steps of implementation
12IFLA Library Reference Model 13Virtual International Authority File 14Shapes Constraint Language
The first step is the necessary preprocessing of data sets, dealing with messy data in the world of libraries is very important, [22] described the efforts to deal with multiple Date formats in a dataset, we used OpenRefine [25], [26], which provided the ability to cluster similar names and correct mistakes semi-automatically, then applied reconciliation for author names, Titles, publishers, etc., so we obtained the URIs of these entities according to Wikidata and VIAF, to enrich our local datasets with LOD, using Natural Language Processing here could be very useful, for example to eliminate some words that are usually used as titles for author like(
)
In this stage we couldn’t find unified automated Authority files in Arabic, no automated Arabic subject heading lists, no automated Arabic thesauri for libraries, and when trying to use the available English subject heading such as LCSH, librarians in local libraries assured that they are difficult to use and not suitable for Arabic subjects, which means there is a great need to develop new Arabic domain ontology for libraries.
The second step is mapping between our datasets and the ontology to generate RDF triples, we used Protégé+ Ontop plugin which provides the ability to reuse mappings and to generate a virtual graph[13], as for the ontology we reused a subset of BIBFRAME2.0 classes and properties in addition to a subset of DCMI Metadata Terms, our prototype application ‘experimental system’ is a client-server application built on the .NET Framework 4.8 and ASP.NET MVC5, the tool used for its implementation is Microsoft Visual Studio Enterprise 2017, with an additional package of Open Source .Net Library (dotnetRdf) that provide a powerful and easy-to-use API for working with RDF [42].
In the application we used the triple store as a union catalog so we could query about books of the other libraries using SPARQL and we can import the retrieved results and modify them then save them in our local database, which saves a lot of time for the librarians as shown in figure 5.
Figure 3. Using Openrefine reconciliation service for authors names with Wikidata service
Figure 4. Using Ontop Mapping Editor, sample of the mapping rules to create RDF Triples
Figure 5. Getting the books written by" " from all the connected libraries
Figure 6. LOD enrichment, new details about the author " ",birth place ,birth date..
And we enriched our local library with linked open data, by connecting to remote SPARQL endpoints and getting some useful details like birthdate, birthplace and the links to the author’s wikidata page or the author’s VIAF page,
as in Figure 6, and we need to mention that we did not fix the formatting intentionally to show the Sparql results from the remote endpoint, not from the local database.
we experimented using two Datasets, First Dataset is a relational database built using MS SQL Server2019, it contains a collection of Arabic or translated into Arabic books and represents a sample of the database in the Arabic Cultural Center in Damascus, Second Dataset is a sample of ‘Jamalon’ bookstore database, available as MS Excel datasheet, Dataset A =1136 records, Dataset J
=8986 records, Table2 shows that there is low availability of the names of Arabic authors on Wikidata and VIAF, the best result comes from VIAF, we can explain the very low rates of availability on ORCID because of the nature of the data sets that contain old and traditional books in the majority, and Arabic letters are not usually used on ORCID which affected the retrieval of the IDs, results of reconciliation can vary in case of using other samples but still very low.
Through the testing process, we have noticed the following:
The ability to use BIBFRAME2.0 classes and properties without the need to translate into Arabic, because the mapping step is executed by the data analyst and the GUI is already in Arabic so dealing with it can be at the application level and not in direct contact with the
user, which can be acceptable in concerns of bibliographic ontology.
Using the predefined classes, object properties and data properties has some difficulties because of the tables in the relational databases did not match with the classes. so we highly recommend considering the architecture of BIBFRAME when planning to develop new databases, and it can be very acceptable to use DCMI Metadata Terms, with the small and medium local Arabic libraries, as a start point.
Mapping to BIBFRAME2.0 classes and properties was executed in our prototype at the ITEM level to achieve migrating old data while adding new WORK or INSTANCE is available for new entries one by one.
We have taken into consideration the limited resources of some Arabic countries like Syria, Yemen, and Sudan, so open-source tools are mostly used here.
The focus in the Arab library world so far is on MARC and the difficulty of accepting the change easily, which is not limited to the Arab world. Most of international libraries still use MARC while trying to convert to BIBFRAME including the library of Congress, due to the difficulties of conversion of large blocks of data[28]
Service | Author Name | Title | subject | publisher |
Wikidata | 40% | 5% | 15% | 14% |
ORCID | 1% | - | - | - |
VIAF | 61% | 6% | 0% | 10% |
(a)
Service | Author Name | Title | subject | publisher |
Wikidata | 42% | 10% | 27% | 2% |
ORCID | 0% | - | - | - |
VIAF | 65% | 1% | 1% | 0% |
(b)
Table 2. Availability of Arabic metadata as linked open data, (a) represents the first sample with 1136 record,
(b) represents the second sample with 8986 records
Reusing standard Bibliographic Ontologies to achieve interoperability and data integration between Arabic libraries is applicable with good results, while there is a clear weakness in the availability of Subject Ontologies in Ara-
bic and the difficulty of using Subject Ontology of other languages, the availability of open source tools to enrich our libraries with LOD can be of great help, our future work includes developing Arabic library subject ontologies and more focusing on natural language processing techniques.
Abdo, R.Z. (2013) FRBR MARC using the web ontology language. Medicine Science. Al-Quds University.
Patel, A., Jain, S. (2021) Present and future of semantic web technologies: A research statement. International Journal of Computers and Applications, 43, 413–
422. DOI: 10.1080/1206212X.2019.1570666.
IFLA UNESCO Digital Library Manifesto (2011). Paris.
Kruk, S.R., Woroniecki, T., Gzella, A. (2007) [Online] JeromeDL – A semantic digital library JeromeDL – A semantic digital library. Available: https:// www.researchgate.net/publication/221466019.
Gruber, T.R. (1993) A translation approach to portable ontology specifications. Knowledge Acquisition, 5, 199–
220.
Kruk, S.R., B.M., editors (2009). Semantic Digital Libraries. Springer: Berlin.
Nisheva-Pavlova, M., Shukerov, D., Pavlov, P. (2016) Design and implementation of a social semantic digital library. Information Services and Use, 35, 273–284.
Candela, G., Escobar, P. M. (2021) Marco-such-A Shape Expression approach for assessing the quality of Linked Open Data in libraries, IOS Press,, 1. doi: 10.3233/ SW-210441.
Elizarov, A.M., Kirillovich, A.V., Lipachev, E.K. & Nevzoova, O.A. (2022) OntoMath digital ecosystem: Semantic service and mathematical ontologies. In:. Scientific Conference “Scientific Services and Internet” 24th Scientific Conference ‘Scientific Services & Internet –2022,’ 167–178.
El-Labban, N. (2016) The bibliographic ontologies and the bibliographic framework data model: Comparative Study. Cybrarians Journal, 41.
Asfand-E-Yar, M., Ali, R. (2020) Semantic Integration of Heterogeneous Databases of Same Domain Using Ontology. IEEE Access, 8, 77903–77919.
Chaves Fraga, D. (2021) Knowledge Graph Construction from Heterogeneous Data Sources exploiting Declarative Mapping Rules. PhysiologyD., Politécnica de Madrid.
An inventory of tools for converting data to RDF. https:/
/faircookbook.elixir- europe.org/content/recipes/ interoperability/rdf-conversion.html. accessed 09 December, 2022.
Hidalgo-Delgado, Y., Estrada-Nelson, R., Xu, B., Villazon-Terrazas, B., Leiva-Mederos, A., Tello, A. (2017)
Methodological guidelines for publishing library data as linked data. In: Proceedings International Conference on Information Systems and Computer Science, INCISCOS 2017, Institute of Electrical and Electronics Engineers Inc., Mar. 2018, p. 241–246. DOI: 10.1109/ INCISCOS.2017.17
Roy, P., Arora, D. (2011) Social semantic digital library: The future. DESIDOC Journal of Library and Information Technology, 31, 226–233. DOI: 10.14429/ djlit.31.4.1101
Yousef, R. (2012). Semantic Digital Libraries Systems: A Comparative and Analysis Study “AFLI” A peer-reviewed scientific journal issued by the Arab Federation for Libraries and Information, no. 10s.
Raza, M.Z., Warraich, N. (2017) [Online] Impact of Semantic Web Technologies on digital collections of libraries. Available: https://www.researchgate.net/publica-tion/326405579.
Haslhofer, B., Isaac, A., Simon, R. (2018). Knowledge Graphs in the Libraries and Digital Humanities Domain. DOI: 10.1007/978-3-319-63962-8_291-1.
Raza, Z., Mahmood, K., Warraich, N.F. (2019) Application of linked data technologies in digital libraries: A review of literature. Library Hi Tech News, 36, 9–12. DOI: 10.1108/LHTN-10-2018-0067
Leiva-Mederos, A., Senso, J.A., Hidalgo-Delgado, Y. & Hipola, P. (2017) Working framework of semantic interoperability for CRIS with heterogeneous data sources. Journal of Documentation, 73, 481–499. DOI: 10.1108/ JD-07-2016-0091.
Nafis, F., Yahyaouy, A., Aghoutane, B. (2019) Ontologies for the classification of cultural heritage data. In: International Conference on Wireless Technologies, Embedded and Intelligent Systems (WITS). IEEE Publications.
Dorobat, I.C., Posea, V. (2020) Raising the interoperability of cultural datasets: The Romanian cultural heritage case study. In: Lecture Notes in Business Information Processing. Springer Science+Business Media Deutschland GmbH, 35–48. DOI: 10.1007/978-3-030-63396-7_3
Zapounidou, S. (2020). Study of library data models in the Semantic Web environment,” PhD, Ionian University DEPARTMENT. DOI: 10.1007/978-3-030-63396-7_3.
Simões Patrício, H., Cordeiro, M.I., Nogueira Ramos,
P. (2020). From the Web of Bibliographic Data to the Web of Bibliographic Meaning: Structuring, Interlinking and Validating Ontologies on the Semantic Web.
Miller, M. (2022) OpenRefine: An approachable open
tool to clean research data. Geospatial Data and Software Reviews, 170.
Hansen, C., Crowe, S. (2015) The linkable Neil Armstrong: Using BIBFRAME to increase visibility of digital collections. Dublin Core and Metadata Applications. Available: https://drc.libraries.uc.edu/. In: Vol. 2015, p 181–189 [Online].
Kovari, J., Folsom, S., Younes, R. (2017) [Online]. Towards a BIBFRAME Implementation: The bibliotek-o Framework,. Available: ld4p.org.
Jett, J., Kudeki, D., Worhtley, G., Cole, T.W., Downie,
J.S. (2020) Applying BIBFRAME in large-scale digital libraries: The Hathi Trust Research Center’s experience. Proceedings of the Association for Information Science and Technology, 57.
Zapounidou, S., Sfakakis, M., Papatheodorou, C. Integrating library and cultural heritage data models: The BIBFRAME – EDM case In ACM International Conference Proceeding Series. Association for Computing Machinery (Oct. 2014).
Sfakakis, M., Zapounidou, S., Papatheodorou, C. (2020) Mapping derivative relationships from BIBFRAME
2.0 to RDA. Cataloging and Classification Quarterly, 58, 603–631. DOI: 10.1080/01639374.2020.1821856.
Repke, T., Krestel, R. (2021). Interactive Curation of Semantic Representations in Digital Libraries.
Alamri, S.E. (2020) Ontology extraction from an Arabic book. PhysiologyD. Kent State University.
Kareem, S.S. (2014) Image retrieval system based on Arabic ontology. Medicine Science. Al-Quds University.
Al-Zoghby, A.M., Ahmed, A.S.E., Hamza, T.T. (2013)
Arabic semantic web applications – A survey. Journal of Emerging Technologies in Web Intelligence, 5, 52–69. DOI: 10.4304/jetwi.5.1.52-69.
The standard model ‘BIBFRAME’ for resources description and access in web environment: Applications and challenges (2015). https://cybrarians.info/en/about-
bibf.html. accessed 28 January, 2023.
Alkhayat, M. (2019). The exchange and interoperability of data among automated systems:An applied Study at the Central Library of University of Damascus, M.Sc. Damascus University.
Guha, N. (2014) [Online] Semantic digital library services. Available: https://www.researchgate.net/publication/ 228898098.
Zapounidou, S. (2017) Studying conceptual models for publishing library data to the semantic web. In: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Springer Verlag: Berlin, 652–655. DOI: 10.1007/978-3-319-67008-9_63.
Elizarov, A., Khaydarov, S., Lipachev, E. (2017) Scientific documents ontologies for semantic representation of digital libraries. In: Proceedings of the 2nd Russian-Pacific Conference on Computer Technology and Applications. Russky Island: Vladivostok, Russian Federation. Russian Federation, 1–5. DOI: 10.1108/EL-08-2017-0168.
Khan, S.A., Bhatti, R. (2018) Semantic Web and ontology-based applications for digital libraries an investigation from LIS. Electronic Library, 36, 826–841.
Hidalgo-Delgado, Y., Xu, B., Mariño-Molerio, A.J., Febles-Rodríguez, J.P., Leiva-Mederos, A.A. (2019) [Online]. A Linked Data-Based Semantic Interoperability Framework for Digital Libraries,. Available: http:// rcci.uci.cuhttp://rcci.uci.cu
Coladangelo, L.P. (2020) Thesaurus and ontology construction for contra dance:knowledge organization of a North American folk dance domain. Knowledge Organization, 47, 523–542 .
Li, X.F. (2020) Study on book ontology in the digitization process of books. Journal of Physics: Conference Series, 1544. DOI: 10.1088/1742-6596/1544/1/012035.
Mandal, S. (2022) [Online] Integration of linked open data authorities with OpenRefine: A methodology for libraries. Library Philosophy and Practice (e-journal). Available: https://digitalcommons.unl.edu/libphilprac.