Executable platform for managing customizable metadata of educational resources
DOI | https://doi.org/10.1108/EL-04-2017-0079 |
Date | 10 December 2018 |
Published date | 10 December 2018 |
Pages | 962-978 |
Author | Amel Abdyssalam Alhaag,Goran Savic,Gordana Milosavljevic,Milan Tima Segedinac,Milorad Filipovic |
Subject Matter | Information & knowledge management,Information & communications technology,Internet |
Executable platform for managing
customizable metadata of
educational resources
Amel Abdyssalam Alhaag,Goran Savic,Gordana Milosavljevic,
Milan Tima Segedinac and Milorad Filipovic
University of Novi Sad, Faculty of Technical Sciences,
Department of Computing and Control Engineering, Novi Sad, Serbia
Abstract
Purpose –The purpose of this research is to enable dynamic customization of metadata that describes
educationalresources in digital repositories.
Design/methodology/approach –Users need to describe educational resources in digital repositories
according to a user-specificmetadata set. As users generally do not have the skills to customizethe software
application manually, this approach relies on the techniques of model-driven software engineering, which
should allow customizationof the software application programmatically with no need to develop or order a
new softwareapplication. An experiment was conducted to evaluate the proposedsolution.
Findings –A software platform for managingeducational resources described by dynamically extendable
metadata is proposed. The platform enables the creation of data models that are programmatically
transformed to a Web applicationfor the management of educational resources. In this way,users can create
their own modelsof metadata that are relevant in a particular domain.
Research limitations/implications –The solution has been verified by users with technical
knowledge. The appropriatenessof the model should still be explored for domainexperts with little technical
knowledgewho desire to define new metadata in their domain.
Practical implications –The solution can be used for digitalrepositories that store diverse educational
resources.Each resource could be described using metadata thatrelates to the domain the resource belongs to.
Originality/value –Digital repositories standardly describe educational resources using some general
metadata, which are more focused on the physical characteristics of resources rather than their semantics.
The proposed solution introduces custom domain-specific semantics into the description of the resources,
which improvestheir retrieval.
Keywords Metadata, Digital repositories, Educational resources
Paper type Research paper
Introduction
With the increasing presence of digital educational resources, the quality of the learning
process directly relates to the availability of these resources, as well as to the quantity and
expressiveness of the metadata used for describing them. The availability of educational
resources is mostly providedby storing these items or their metadata in digital repositories.
The repositories usually provide for the searching, displaying and editing of educational
resources. Metadata is an important factor for the searchability of learning objects, as the
information provides additionaldescriptions of the learning objects. These descriptions can
be related to the physical characteristics (e.g.format and length), classification of the object
and the semantics of the learningobject.
Metadata can be divided into two categories. The first category consists of metadata
describing the object’s characteristics that are not related to the domain the object belongs
EL
36,6
962
Received8 April 2017
Revised12 October 2017
6 December2017
Accepted18 January 2018
TheElectronic Library
Vol.36 No. 6, 2018
pp. 962-978
© Emerald Publishing Limited
0264-0473
DOI 10.1108/EL-04-2017-0079
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/0264-0473.htm
to. These metadata are general and can be applied to all learning objects regardless of
domain or discipline. Examples of such metadata are file format, author, language and so
forth. The second category refers to metadata that uses domain-specific information to
describe learning objects. In many domains, classifications have been developed that
categorize content within a specific domain.As an example of the domain-specific metadata
sets, the authors will briefly illustrate the management of educational resources in the
computer science domain. When storing educational resources related to this field, the
search functionality can be improved if the resources are described using domain-neutral
metadata, such as the languageof the educational resource.In addition, it would be useful to
specify the subfield of computer science that the educational resource belongs to. For
example, if the resource explains memory management in operating systems, the authors
recommend using a domain-specific metadata set that contains elements with the
vocabulary, such as software,operatingsystems and memory management.
In general, software systems for managing educational resources face two challenges.
First, systems should support a general metadata set to allow for management of
educational resources that belongto different domains. It is necessary to describe parts of a
resource’s semantics that are domain-specific. However, it can be quite difficult to
implement a software application that contains predefined metadata sets for various
domains. Even if such an application with various initially supported metadata sets was
implemented, there is still the problem of adding new metadata sets that are user-specific.
With statically defined metadata sets, an application would not allow modifications of
existing metadata sets or describing the semantics of educational resources in an entirely
new domain.
The goal of this research is to enable dynamic customization of metadata sets that a
software application for the management of educational resources could support. The main
idea of the research is allowing users to define metadata sets on their own after the
application is developed. Such metadata is referred to as customizable metadata. Users can
customize the application for describing educational resources’semantics in their domain.
As it takes much effort to develop a completely new application for managing educational
resources according to their models, theresearchers developed an executable platform that
generates the final software application programmatically. This solution ensures that users
will be able to manage educational resources using the semantics from the specific domain
with no need to develop or order a new software application. The executable platform
provides a dynamic adaptationof metadata sets to the required domain.
The authors present an executable platform for generating a software application that
provides management of educational resources described by customizable metadata sets.
The platform is an extension of a Kroki prototypingtool (Kroki tool, 2014), which has been
built on the principles of model-drivenengineering.
Literature review
This section presents principles and specifications that the research relies on and presents
other approaches aimed at similar goals. To begin, different software systems, which
provide availability and managementof educational resources, will be analysed. Numerous
systems, commonly named as learning object repositories, have been developed for this
purpose. These systems are mostly implemented as public internet repositories that store
educational resourcesand/or related metadata.
MIT OpenCourseWare (MIT, 2001) is a repository that allows free downloading of
learning material used in MIT courses. Similarly, Carnegie Mellon University (CMU) within
its project of the open learning initiative (OLI) (CMU, 2001) provides open access to the
Managing
customizable
metadata
963
To continue reading
Request your trial