Building Cloud Data Interchange Services for E-Learning Systems:

Applications on the Moodle System

Alina Andreica

and Fernando Paulo Belfo

Babes-Bolyai University, Cluj-Napoca, Romania

Instituto Politécnico de Coimbra, Coimbra, Portugal

Keywords: Learning Components, Learning Objects, Equivalence and Simplification Algorithms, Data Interchange,

Cloud Services, Database Representation, Pattern Matching, Software Design.

Abstract: The cloud data interchange services among various information systems still have huge possibilities to

evolve. We focus on applying data interchange principles that we have proposed for cloud environments in

order to perform data exchange for e-learning systems. Our case study is based on the moodle system.

Developing and improving data interchange standards for learning objects contributes to improving

processing and exchanging techniques of digital content used for teaching, learning, or training among

various e-learning systems. Equivalence algorithms and canonical representation are used in order to ensure

the uniform representation in the cloud database. The solution we describe, designed for cloud architectures,

has important advantages in e-learning systems communication and educational institutions cooperation,

since different institutions, using different e-learning systems, do not have automatic means of exchanging

learning information. Therefore, these techniques for representing and exchanging learning objects facilitate

their sharing among different stakeholders.

1 INTRODUCTION AND

WORKING FRAMEWORK

Developing standards for representing and

exchanging learning resources among different

stakeholders involved in educational processes bring

them many practical advantages. Content authors

may want their work to be easily spread and used, or

they may want to collaborate with other co-authors

on the same content (Andreica, Synasc 2017).

Beneficiaries may want to be able to easily choose,

render, or buy learning content, in order to learn

more effectively. Standardizing the way in which

learning objects are represented contributes to

increasing their availability, accessibility, in a way

they can be valued and, therefore, they can be easily

transacted between potential beneficiaries and

providers.

In order to exchange learning resources between

different parties, the information systems have to be

interoperable. Interoperability is the capability of

different systems to share functionalities or data

(Olmedilla et al, 2006). System interoperability has

been approached on business models (Ziemann,

2010) and other various models (Morris et al, 2004).

Andreica, Covaci and Küng overviewed

interoperability layers, principles and tools

(Andreica et al, 2015), including: the Electronic

Data Interchange (EDI) model (Adams et al, 2002),

the XML standard (XML standard, 2015),

RosettaNet (Rosetta, 2015), ebXML (OASIS, 2015)

standards. Standards for e-learning objects and

components are given in

(SREB, 2007), (IMS,

2010). The Open Internet of Things standards

(OpenIoT, 2015) may also be used as an efficient

framework for data interchange (Andreica, Synasc

2017).

A widely used definition for the learning object

(SREB, 2007) is the one proposed by the Institute of

Electrical and Electronics Engineer’s (IEEE): “any

entity, digital or non-digital that can be used, re-used

or referenced during technology supported learning”

(IEEE, 2002). This definition is a very

comprehensive one, anchored on the idea of

“technology supported learning”, yet a bit too large

to be practically used for learning objects or

components within e-learning system environments.

Andreica, A. and Belfo, F.

Building Cloud Data Interchange Services for E-Learning Systems: Applications on the Moodle System.

DOI: 10.5220/0006783905650572

In Proceedings of the 8th International Conference on Cloud Computing and Services Science (CLOSER 2018), pages 565-572

ISBN: 978-989-758-295-0

565

More e-learning definitions are discussed in section

Learning objects can have different formats and

can either consist of simple assets as text, images or

short videos, or can comprise more complex

resources, with several components, combined in

order to deliver meaningful learning experiences

(Andreica, Synasc 2017). These components may

consist of combinations of text, images, simulations,

videos, assessment exercises or several other ways

of achieving an educational objective (IMS, 2010).

In order to handle these composition aspects, the

SREB-SCORE standard defines two levels of

learning items (SREB, 2007): the first level is

composed by assets and information objects and the

second level is composed of learning objects and

learning components. Assets are electronic

representations of media, text, images, sounds, web

pages and other reusable pieces of data. An

information object is composed of assets that focus

on a particular part of information, which can

illustrate a principle, explain a concept or describe a

process. The learning objects cover assets and

information objects that teach a single concept or

lesson (Andreica, Synasc 2017). The components of

learning objects are their objectives, assessments,

practices, concepts, but also their lessons, modules

and courses (SREB, 2007).

We use equivalence algorithms in order to set

the canonical representation of the exchanged data in

the cloud database (Andreica, 2017).

The current paper applies the proposed

framework for using simplification and equivalence

algorithms in modelling data representations

(Andreica, 2017) for learning objects and

components within the case study performed on

moodle system. We apply these techniques in

designing data interchange service among various

information systems, in particular e-learning

systems, within cloud environments.

Within section 2 we overview the means we have

proposed for implementing simplification and

equivalence algorithms on various entities, including

hierarchical structures and techniques for solving

specific pattern matching problems using canonical

representatives. Within section 3, we focus on e-

learning objects and components which appear in the

two moodle e-learning system instances that we use

as case studies. Section 4 addresses principles of

data interchange between information systems using

a cloud database, which retains data in a canonical

representation. We describe the structures for the

entities and attributes to be used in the local and

cloud databases and we discuss the results of

transferring course related tables between different

moodle databases. Conclusions reveal the most

important topics presented in the paper and future

research and development directions.

2 EQUIVALENCE ALGORITHMS

AND PATTERN MATCHING

PRINCIPLES

Within this section we overview the methods we

propose (Andreica et al, 2010), (Andreica et al,

2012) for implementing equivalence algorithms at

the database level for various entities and entity

classes, including hierarchical structures, the entities

being retained in database tables. This solution is

very useful, since we often have to process large

volumes of data retained in databases.

The equivalence algorithms we have

implemented are based on the theoretical framework

given in (Buchberger and Loos, 1982).

Principles for retaining (using ascendant /

successor pointers) and processing hierarchical

structures, postorder type n-ary tree parsing

algorithms and a comparison of their processing

techniques are presented in (Andreica et al, 2010).

We prove that the function implementing the

equivalence test for two entities based on the

property that they have the same canonical

representative is much more efficient than the one

using the definition (Andreica et al, 2010).

In (Andreica et al, 2012) we discuss the case

study of equivalent disciplines and processing

modules of didactic activities. In (Andreica, 2016)

we approach the case of an expert system for plant

therapy. In (Andreica, 2017) we propose principles

for designing uniform database structures in order to

build cloud data interchange services between

information systems. In (Andreica, Synasc 2017) we

add the framework of e-learning systems.

The principles for implementing equivalence

algorithms on categories of entities are also

described in (Andreica et al, 2010), (Andreica et al,

2012). Intuitively, the canonical set of a category of

entities may be obtained by “flattening” its category

sub-tree and computing the union set of all canonical

sets of its descendant leaf entities. In the case of

categories of entities, the canonical representative is

recursively computed (Andreica, 2017).

In many practical cases, entity equivalences or

even canonical elements have to be mapped,

sometime with user assistance. Therefore, in

(Andreica, 2016) and (Andreica, 2017), we address

CLOSER 2018 - 8th International Conference on Cloud Computing and Services Science

566

mappings and pattern matching issues. We overview

here the principles we proposed in this respect.

We designed pattern matching rules for

equivalent entities by reducing the mapping between

two elements, belonging to two equivalence classes

that are to be mapped, to mapping their canonical

representatives (Andreica, 2016) and (Andreica,

2017). The formalization is given in (Andreica,

2016), (Andreica, 2017).

Intuitively, instead of dealing with the

equivalence of any elements between two classes,

we reduce the problem to the simpler one of dealing

with the equivalence of the two classes’

representatives – see figure 1.

Figure 1: Pattern Matching scheme for Equivalence

classes (Andreica, 2017).

For the case of equivalence classes with hierarchical

representations, the canonical representatives of the

classes are the roots of the corresponding trees

(Andreica, 2017)

3 LEARNING RESOURCES,

OBJECTS AND COMPONENTS.

CASE STUDIES FOR MOODLE

E-LEARNING SYSTEMS

E-learning is defined by Clark and Meyer as

“instruction delivered on a digital device such as a

computer or mobile device that is intended to

support learning” (Clark & Mayer, 2016). The IEEE

perspective on learning objects views them as

entities, digital or not, which can be used during

technology supported learning, this technology

including or not computers or mobile devices.

We consider that a more digital and computer

targeted definition for learning objects, would better

support the e-learning field. In this respect, the

definition proposed by David Wiley, as “any digital

resource that can be reused to support learning” is

more applicable for the e-learning (Ritzhaupt, 2010;

Wiley, 2000), since it views the learning objects as

digital ones, which and be delivered across any

network.

In the e-learning field it is important how

information and communication technologies are

used to facilitate learning, especially over the

Internet, the management of learning objects being

therefore an important task.

We further present types of learning resources

within two moodle case studies systems, the final

goal being to discuss means of exchanging learning

objects. Moodle, a “learning platform designed to

provide educators, administrators and learners with a

single robust, secure and integrated system to create

personalized learning environments”. Moodle is an

open source software, provided freely under the

GNU General Public License (Moodle, 2016).

The first case study system is used within the

Institute of Accounting and Administration in

Coimbra, Portugal, one of the schools of Polytechnic

Institute of Coimbra. Moodle platforms manages

different types of learning objects (see figure2),

providing services for creating and uploading

learning objects.

The types of objects displayed in the figures

below are to be mapped using the mapping

procedures discussed in section 4.

Figure 2: Types of learning objects in moodle.

Figure 3 and 4 present respectively features of

the moodle platform regarding browsing repositories

by course titles (figure3) and listing educational

resources (figure 4).

Yet, as it is publically stated on the Moodle

official site, there are several issues which are not

yet solved and have to be discussed in the future, for

example, metadata, tracing student learning and

searchable repository services.

Building Cloud Data Interchange Services for E-Learning Systems: Applications on the Moodle System

567

Figure 3: Browsing repositories by course title.

Figure 4: Educational resources listed in ISCAC Moodle.

The second system is a moodle test instance

installed within Babes-Bolyai University in order to

test data transfer between the two systems.

Within this system, we have created an IT

strategies course both for educational purposes, and

for data exchange ones. Figures 5-6 present the way

in which educational resources have been added

within this course. Unlike previous example,

educational files have been added as links.

Figure 5: Course topics in design mode, Moodle instance -

Babes-Bolyai University.

Figure 6: Course topics in reading mode, Moodle instance

- Babes-Bolyai University.

4 PRINCIPLES FOR BUILDING

THE CLOUD DATABASE

REPRESENTATION AND FOR

TRANSFERING DATA

We have previously addresses the topic of building

the cloud database representation in (Andreica,

2016), (Andreica, 2017), (Andreica, Synasc 2017).

Within this section, we overview the previously

stated principles, including data structuring

principles, and give further details on applying them

on learning objects and components.

The cloud database retains the canonical

representation of the exchanged data and is used as a

communication proxy. The exchanged data consists

of learning objects as the ones presented in section

3. We have designed the cloud database as a

communication proxy in order to increase

communication efficiency and to retain the

canonical representation of the exchanged data.

Therefore, if systems S

and S

have communicated

and S

have also communicated, then their

mappings to the canonical representation have

already been performed, consequently S

and S

can

communicate straightforwardly, without any

supplemental mappings (Andreica, Synasc 2017).

The mapping process is user assisted, requiring

human input.

The communication principle previously stated is

also applied in the case of the systems presented in

section 3 – the Moodle system of Coimbra Business

School and the Moodle test instance installed within

Babes-Bolyai University. Moreover, we have also

created a blank moddle database in the test instance

installed within Babes-Bolyai University in order to

test table transfer.

CLOSER 2018 - 8th International Conference on Cloud Computing and Services Science

568

The data exchange sequences between a source

system Ss and a destination system Sd via the cloud

database Scd pursue the following stages (Andreica,

2017): data to be exchanged is marked in the Ss

database using dedicated tables and columns,

mapped into the Scd database, and then sent and

retained into Scd. For the destination system

database Sd, which sends data requests, a data

mapping is also performed to Scd and required data

is sent from the Scd into the destination one Sd. The

exchange framework is graphically represented in

figure 7 (Andreica, 2017).

Figure 7: Data Interchange Model Using Cloud Services

(Andreica, 2016).

The data exchange uses multi-criteria agents

(Weiss, 1999) implemented in the cloud

environment for performing necessary mappings and

for handling communication (Andreica, 2016). The

agent architecture includes self-adapting

communicating objects, working on distributed

datasets and supporting the exchange and analysis of

distributed sources (Andreica et al, 2015).

In the proposed architecture (Andreica, 2017),

when agents communicate, they may decide to

cooperate on a given task. In this respect, they make

a collective commitment to each other. Committing

to another agent involves agreeing to pursue the

given goal-data interchange (Andreica, 2017). The

agent architecture is based on the multi-agent model

proposed in (Faulkner et al, 2014)

In (Andreica, 2017) we propose a database

model and structure for processing the entity

equivalence from tables in various local databases,

using the uniform canonical representation in the

cloud database. The formalization of the local and

cloud database structures and their mapping are

given in (Andreica, 2017).

The database structures we propose in the local

and cloud databases are described in details in

(Andreica, 2017), (Andreica, Synasc 2017).

Learning objects are retained for each learning

component, according to its characteristics. They

may usually include: files, texts, images, links,

documents (Andreica, Synasc 2017).

One of the most important learning component

is the moodle lesson or learning module. In the

examples given in section 3, lessons / topics in

Moodle contain the following objects: Title – text;

Goal – text; Learning materials – list of learning

files or links to them; References – text (may include

links); Evaluation information – text, learning file or

link to a file. In the ISCAC moodle system,

evaluation information is included in the learning

materials.

In order to prepare the testing framework, we

have installed a moodle test instance within Babeş-

Bolyai University. In figure 8 we display the

communication tables from the local database in this

moodle test instance 193.231.19.229/moodle .

Figure 9 presents the structure and content of the

moodle block recent activity table.

We have studied the Moodle database schema

(Moodle DBS, 2017) (Moodle Courses Schema,

2017). The course component of this schema is

presented in figure 10. Further on, we have

transferred the moodle block recent activity table

into a blank moodle database, called moodletest,

using an intermediate XML structure, namely the

table conversion to XML and vice-versa – see figure

11. In (Prasad, 2017), Prasad proposes the SQL code

sequence for transferring a table into the

corresponding XML structure and from the XML

structure into the corresponding table. In our

example, into the new table,there were inserted 25

rows and the query took 0.0636 seconds – see figure

11.

5 CONCLUSION AND FUTURE

WORK

The paper applies the data representation and design

principles we have proposed for performing data

interchange between various information systems

databases, by means of cloud services, for e-learning

systems, the case studies being implemented on

moodle systems.

Cloud

service

Cloud Generic

Database

Information

stem 1 DB

Information

system 2 DB

Information

sys

m n DB

Building Cloud Data Interchange Services for E-Learning Systems: Applications on the Moodle System

569

Simplification and equivalence algorithms are

used in order to ensure canonical data representation

in the cloud database and data mapping in the data

exchange process. A specific structure for the

entities / tables and attributes that are to be

exchanged is used in the local and cloud databases.

Pattern matching rules and canonical representatives

are used in the cloud database.

We reveal the advantages of applying the

algebraic equivalence algorithm and canonical

representatives’ properties in designing data

interchange services for various information system

cases, the present paper focusing on e-learning

systems, with case studies on moodle system.

The data interchange model we have developed

provides important practical advantages for

increasing organizational competitiveness in the

educational field, as well as between educational

actors and other stakeholders, with a significant

societal impact on institutional cooperation and

improvement of learning processes, efficient

information access, learning and communication for

various stakeholders.

The solution we have designed is flexible and

efficient, only relevant data being exchanged,

minimizes resource usage and has significant

security benefits.

Future work is related to further development

and implementation of the above described

techniques on e-learning systems and in other fields

as well.

Figure 8: Communication tables in the local moodle database.

Figure 9: Moodle block recent activity table, test instance http://193.231.19.229/moodle.

CLOSER 2018 - 8th International Conference on Cloud Computing and Services Science

570

Figure 10: Moodle course tables; Source: https://docs.moodle.org/dev/Database_schema_introduction#Courses_and_their_

organisation_into_categories

Figure 11: Table transfer for Moodle block recent activity table by means of an XMLstructure inti moodletest datadase.

REFERENCES

Andreica, A, Stuparu, D, and Miu, C., 2012. Applying

Mathematical Models in Software Design, 2012 IEEE

8th International Conference on Intelligent Computer

Communication and Processing, Cluj-Napoca,

Romania, Proceedings of ICCP 2012, IEEE, Ed: Ioan

Alfred Letia, p.87-90

Building Cloud Data Interchange Services for E-Learning Systems: Applications on the Moodle System

571

Olmedilla, D., Saito, N., and Simon, B. eds. (2006).

“Educational Technology & Society”, Special Issue on

Interoperability of Educational Systems, vol 9

Andreica, A, Stuparu, D, and Miu, C., (2010). “Design

Techniques in Processing Hierarchical Structures at

Database Level”, Proceedings of Iadis Information

Systems 2010, Porto, 18-20 March 2010 , IADIS

Press, Ed: M Nunes, P Isaias, P Powell, p. 483-488

Andreica, A, Covaci, F. and Küng, J. (2015). A Generic

Model for Cloud Data Interchange, Proceedings of

14h RoEduNet International Conference - IEEE,

Craiova, 24-26 September 2015, IEEE Computer

Society, p. 138-142

Andreica A (2017) Designing Uniform Database

Representations for Cloud Data Interchange Services,

7th International Conference on Cloud Computing and

Services Science, 24-26 April 2017, Porto, Portugal,

Proceedings of CLOSER 2017, SCITEPRESS –

Science and Technology Publications, Editor: Donald

Ferguson, Víctor Méndez Muñoz, Jorge Cardoso,

Markus Helfert and Claus Pahl, P. 554-559

Andreica, A, Belfo F P, Covaci F., (2017) Designing Data

Interchange Services for Learning Components, paper

presented at Symbolic and Numeric Algorithms for

Scientific Computing Synasc 2017

Adams, S, Hardas, D, Iossein, A. and Kaiman,. C. (2002)

BizTalk Unleashed. Indianapolis, Indiana: Sams

Publishing. p. 966

XML standard, http://www.w3.org/TR/xml11/#charsets ,

retreived Nov 2015

Rosetta – Rosettanet Overview: Clusters, Segments, and

PIPs (ver 02.13.00), 2011, retrieved December 2015

http://www.rosettanet.org/TheStandards/RosettaNetSt

andards/PIPOverview/tabid/3482/Default.aspx

OASIS – OASIS ebXML Messaging Services Version 3.0:

Part 1, Core Features, 2007. http://docs.oasis-open

.org/ebxml-msg/ebms/v3.0/core/ebms_core-3.0-spec

.pdf, retreived Dec 2015

Stumme, G., Wille, R. (2000). „Begriffliche

Wissensverarbeitung / Conceptual Knowledge

Processing“, Springer Verlag,

OpenIoT - Open Internet of Things architecture https://

github.com/OpenIotOrg/openiot/wiki/OpenIoT-

Architecture , retreived Dec 2015

Buchberger B. and Loos, R. (1982) Algebraic

Simplification, Computing, Suppl. 4, Springer Verlag,

p.11-43

Andreica, A (2016) Applying Equivalence Algorithms in

Solving Pattern Matching Problems. Case Study for

Expert System Design, Proceedings of International

Conference on Theory and Practice in Modern

Computing- TPMC, July 1-4, 2016, Portugal

Andreica, A, Stuparu, D. and Mantu, I. (2005). “Symbolic

Modelling of Database Representations”, International

Symposium on Symbolic and Numeric Algorithms for

Scientific Computing 2005, IEEE Press, p 59-62

Morris, E., Levine, L, Meyers, C., Place, P. and Plakosh.

D. (2004) “System of Systems Interoperability (SOSI):

Final Report”, Carnegie Mellon Univ., Software

Engineering Institute, http://www.sei.cmu.edu/reports

/04tr004.pdf , accessed May 2016

Weiss, G. E, (1999). “Multiagent Systems: A Modern

Approach to Distributed Artificial Intelligence”, MIT,

Faulkner, S, Kolp, M, Nguyen, Tai, Coyette, A, Do, T.,

2014. “Information Integration Architecture

Development: A Multi-Agent Approach”, retr Nov

2016

Clark, R. C., & Mayer, R. E., 2016. E-learning and the

science of instruction: Proven guidelines for

consumers and designers of multimedia learning

(Third ed.). San Francisco: Pfeiffer.

Harman, K., & Koohang, A., 2007. Learning objects:

Applications, implications, & future directions. Santa

Rosa, California: Informing Science Press.

IEEE, 2002. 1484.12.1 Learning Object Metadata

Standard.

IMS, 2010. IMS Learning Object Discovery & Exchange

(LODE). IMS Global Learning Consortium. 2017,

from https://www.imsglobal.org/lode/index.html

Moodle, 2017. Moodle Documentation. Retrieved July 16,

2017, from https://docs.moodle.org/33/en/About_

Moodle, https://docs.moodle.org/23/en/Activities

Moodle Courses Schema, 2017. Moodle Courses

Database Organization, retreived October 2017 from

https://docs.moodle.org/dev/Database_schema_introdu

ction#Courses_and_their_organisation_into_categorie

Moodle Database Schema, 2017. Retrived October 2017

from http://www.examulator.com/er/

Prasad, D, 2017. How to convert table data to XML and

vice-versa, retreived October 2017 from

http://sqlserverlearner.com/2013/how-to-convert-

table-data-to-xml-and-vice-versa

Ritzhaupt, A, 2010. Learning Object Systems and

Strategy: A description and discussion.

Interdisciplinary Journal of e-learning and Learning

Objects, 6(1), 217-238.

SREB, 2007. Learning Object Metadata - SREB-SCORE

Initiative. Atlanta: Southern Regional Education

Board.

Wiley, D. A, 2000. Connecting learning objects to

instructional design theory: A definition, a metaphor,

and a taxonomy. In David A. Wiley (Ed.), The

instructional use of learning objects: Online Version.

Ziemann, J., 2010. “Architecture of Interoperable

Information Systems - An Enterprise Model-Based for

Describing and Enacting Collaborative Business

Processes”, Logos Verlag, Berlin

CLOSER 2018 - 8th International Conference on Cloud Computing and Services Science

572