A Cloud-based Framework for Personalized Mobile Learning

Provisioning using Learning Objects Metadata Adaptation

Elarbi Badidi

College of Information Technology, UAE University, P.O. Box. 15551, Al-Ain, U.A.E.

Keywords: Mobile Learning, Adaptive Learning, Cloud Computing, Learning Objects.

Abstract: With the proliferation of Internet-capable mobile handheld devices and the availability of wireless

broadband networks, mobile learning is increasingly adopted to deliver learning content anywhere and

anytime to mobile users. Offering compelling mobile learning solutions faces several challenges. These

challenges are mainly the adaptation of the learning material to the profile and preferences of the mobile

user and the support of multiple devices. Other concerns include the storage, retrieval, and processing of

learning content outside of mobile devices. Furthermore, building rich learning management systems

requires the integration of learning content from third party providers. This paper describes our proposed

cloud-based framework for delivering adaptive mobile learning services. The paper explains the benefits

and requirements of cloud-based solutions for educational organizations, and describes the components of

the proposed framework together with the process of integrating learning objects imported from third-party

providers with in-house learning objects of the educational organization.

1 INTRODUCTION

E-learning continues to grow phenomenally in both

academia and industry, but most e-learning

developments involve wired infrastructures. In parallel

with that trend, we are witnessing a growing interest in

mobile learning (or m-learning) solutions, which are

fuelled by the proliferation of modern handheld devices

(such as smartphones and tablets) having advanced and

sophisticated technological capabilities and the

availability of wireless broadband connections.

Mobile learning is relatively immature with

respect to technologies and principles and methods

of instruction, but it is evolving rapidly. A review of

the literature reveals several initiatives on the

adoption of mobile learning. Mobile handheld

devices are supporting training of corporate mobile

workers (Pimmer et al., 2008) and are enriching

medical training (Davies et al., 2012), and music

composition (Jung et al., 2006).

Traditional research-based learning systems fall

into two categories: Intelligent Tutoring Systems

(ITS) and Adaptive Hypermedia Systems (AHS)

(Brusilovsky, 2003). AHS solutions focus mainly on

the adaptation of the instructional process (course

content adaptation, course navigation adaptation,

problem-solving support, etc.) to the learner model.

Commercial Learning Management Systems (LMS),

like Blackboard and WebCT, focus mainly on the

management of the learning process (registration

and tracking of students, learning material creation

and delivery capability, skill assessment,

communication teacher-learner and learner-learner,

etc.). The educational and corporate training market

continue to adopt mainly these systems, which LMS

companies developed in the context of classical

learning using wired infrastructure.

This work aims to take advantage of the

proliferation of mobile devices and the promises of

cloud computing to propose a mobile learning

solution. We present in this paper a cloud-based

mobile learning framework whose main objective is

to fill the gap between the current approach to Web-

based education (based on mobile devices, wireless

networks, and learning object repositories), and

robust but underused ITS and AHS technologies.

This framework attempts to address both the

adaptation of the learning material to the mobile

learner model, the user mobility, the heterogeneity

of mobile devices, and the integration of external

learning material from various providers and

learning object repositories (Lehman, 2007). The

motivation behind considering a cloud-based

solution for mobile learning is that mobile devices

are still lacking the necessary resources regarding

368

Badidi, E.

A Cloud-based Framework for Personalized Mobile Learning Provisioning using Learning Objects Metadata Adaptation.

In Proceedings of the 8th International Conference on Computer Supported Education (CSEDU 2016) - Volume 1, pages 368-375

ISBN: 978-989-758-179-3

storage and computing power, compared with a

conventional information-processing device such as

a workstation or a laptop.

The remainder of the paper is organized as

follows: Section 2 presents background information

on mobile learning. Section 3 presents pertinent

literature and related works on adaptive mobile

learning. Section 4 describes the benefits of cloud

computing for mobile learning. Section 5 presents

the architecture of our proposed framework. Section

6 details the process of integrating learning objects

from third party providers. Finally, section 7

concludes the paper and highlights future work.

2 MOBILE LEARNING

As a result of the improved capabilities of mobile

devices and the increasing availability of wireless

networks, there are great opportunities for using

mobile learning as a new channel to convey

knowledge and complement the already established

Web-based e-learning model.

Numerous studies have defined mobile learning

differently, which suggests that mobile learning is

still in an evolving stage. Traxler J. (Traxler, 2005)

defined mobile learning as: “Any educational

provision where the sole or dominant technologies

are handhelds or palmtop devices.” Crompton et al.

(Crompton, 2013) defined mobile learning as:

“learning across multiple contexts, through social

and content interactions, using personal electronic

devices”. Gipple et al. (Gipple, 2016) made a

blurred distinction between mobile learning and e-

Learning: “mLearning facilitates learning ‘on the

move’, and so it is not static or tied to a place; it is

learning within context. This may include learning

while traveling, driving, sitting, or walking; it may

be hands-free learning or eyes-free learning. These

interpretations impact on the implementation and

hence the definition of mobile learning. By focusing

on the context how mLearning is being used,

designers and developers can identify the

advantages that mobile learning can provide for

learners rather than on the limitations of the

technology. It allows educators to capitalize on

learner engagement inherent in mobile technology

to provide learners with the best and most

convenient learning tools possible.”

Two main characteristics of mobile learning are

ubiquity and mobility. Ubiquity represents the state

or capacity of accessing computing technologies and

learning material whenever and wherever the mobile

learner needs them. Mobility represents the quality

of being able to learn while on the go using various

mobile handheld devices. Mobile learners can access

learning services from anywhere, and anytime.

Other characteristics of mobile learning include:

• Interactivity: mobile learners manage the learning

process they are involved in as opposed to the

traditional learning in which learners sit passively

while the instructor feeds them with information.

• Ability to Access a Variety of Learning Material

Anytime from Anywhere, which can help in

understanding the learning concepts under study.

• Flexibility: mobile learning is spontaneous and

not planned in advance.

• Collaboration: mobile technologies in addition

to social networking technologies provide

opportunities for collaboration between learners

themselves and collaboration with instructors.

Some case studies and projects investigated and

experimented with mobile learning as a new channel

to convey knowledge. Cavus et al. (Cavus, 2009)

investigated the potential of using wireless

technologies in learning new technical English

words. Results of the study, in which forty-five

students participated, showed that the students

learned in an effective way new technical

vocabulary using their cell phones. 9ine Consulting

(Heinrich, 2012) investigated the use of iPads at

Longfield Academy in Kent in which over 800

students had iPad across all levels of the school. The

study revealed the value of the iPad as an

educational tool, the involvement of the teachers, the

motivation of the students in using the iPad, and the

rising progress in the quality of students’ work.

3 RELATED WORK

3.1 Adaptive Learning

Several research efforts widely investigated adaptive

learning and personalization of learning material

according to the learner’s model in web-based

learning systems. Personalization of the learning

material has been studied and evaluated in the areas

of psychology of learning and teaching methods

(Tennyson, 1988), (Litchfield, 1990). The empirical

evaluation of these methods showed that

personalized course material increases the learning

speed and help learners understand better the

teaching material (Brusilovsky, 2003).

Well-known projects include AHA (De Bra,

2001) (De Bra, 2002), DCG (Vassileva, 1998), and

ELM-ART (Weber, 2001). AHA is a generic system

for adaptive hypermedia whose aim is to bring

A Cloud-based Framework for Personalized Mobile Learning Provisioning using Learning Objects Metadata Adaptation

369

adaptivity to web-based applications. It supports

adaptive content presentation and adaptive

navigation. DCG is an authoring tool for adaptive

courses. It supports adaptive sequencing and offers

different levels of re-planning the course. ELM-ART

is an on-site intelligent learning environment that

supports example-based programming, intelligent

analysis of problem solutions, and advanced testing

and debugging facilities.

Within the context of mobile learning, only few

research works have investigated the issue of

adaptation and personalization of learning content

based on the learner’s model. Economides et al.

(Economides, 2006) described a general framework

for adaptive mobile learning. The input to the

adaptation engine is the learner’s state, the

infrastructure’s state, the current educational

activity’s state, and the environment’s state. Each

one of these states consists of some dimensions.

Probabilistic adaptation decisions are made when

context information is inaccurate. Jung et al. (Jung,

2006) proposed a mobile learning system that adapts

content provided to the learner based on her

attributes. The adaptation engine relies on a user

model, a domain model, and an adaptation model as

in AHA (De Bra, 2001) (De Bra, 2002). The user

model is shaped as a collection of pairs of

attributes/values. The domain model is a hierarchical

structure of learning content. The adaptation model

consists of a set of rules on how to update the user

model based on the progress in learning. Fang et al.

(Fang, 2009) proposed an architecture for adaptive

mobile learning, which relies on a learning object

model. Learning Objects represent digital resources

that are reusable. The authors of this work provided

a taxonomy of the learner model, of the learner’s

environment model, and the learning object model.

However, they did not provide a detailed

representation and description of each model. They

described the adaptation process only at the

conceptual level. Other works typically implemented

the adaptation process as a set of adaptation rules.

Huang (Huang et al., 2012) proposed an approach to

transform HTML-based content into formats

appropriate for mobile devices. The approach relies

on the concept of coherence and an algorithm that

detects coherence sets. The drawback of this

approach is its limitation to HTML content.

Learning systems often use multimedia content that

requires various technologies for its presentation.

3.2 Cloud-based Mobile Learning

Cloud-based mobile learning is becoming the

subject of several research efforts. Velev (Velev,

2014) described some of the challenges and

opportunities in the development and use of cloud-

based mobile learning. He described how social

media, big data, and cloud computing could make

possible the development of mobile learning

solutions through mobile devices, allowing learners

to access learning content while on the move. Masud

(Masud, 2013) proposed a high-level conceptual

cloud-based architecture to support mobile learning.

The proposed solution advocates a private cloud that

could be used by higher education institutions.

However, the authors did not describe the

interactions between the components of the

architecture from the learning perspective. They just

described the technical details of their solution,

which includes servers, virtualization, storage, and

network access.

The main cloud providers and technology

companies such as Google, Microsoft, and CISCO

are promoting their solutions for cloud-based

education. For example, Google Apps for Education

offers email, calendar, website creation and office

applications, communications, to every student and

staff member in the educational institution.

When building mobile learning solutions, the

following technical issues need to be considered:

a) How to implement and manage the adaptation of

the learning material to the learner’s preferences

and profile?

b) How to provide support for multiple mobile

devices?

c) How to extend the solution with the learning

material from third party providers?

In this work, we propose a mobile learning

solution based on cloud-based services and a two-

fold student modeling mechanism. Cloud-based

services address the problem of managing the

learning infrastructure and multi-device support and

adaptation. The two-fold student modeling allows

handling the personalization process in a very

flexible manner.

4 CLOUD-BASED MOBILE

LEARNING PROVISIONING

Cloud computing enables a service provisioning

model for computing services that relies on the

Internet. This model typically involves the

provisioning of dynamically scalable and virtualized

services. The advent of cloud computing has an

impact on developers, end-users, and organizations.

CSEDU 2016 - 8th International Conference on Computer Supported Education

370

For developers, cloud computing provides

greater amounts of storage than ever before and

better processing power for running the applications

they develop. For end-users, a user using the cloud

through a native application or a web-based

application can access his documents and files

whenever he wants and wherever he is, rather than

having to remain at his desk. Also, cloud computing

opens the door to group collaboration as users from

different locations might share documents and files

at lower costs and in an efficient way. Small and

medium-sized businesses might also benefit

immediately from the huge infrastructure of the

cloud without being concerned with its

administration. They might store massive amounts

of data than on their premises’ systems. Therefore,

their computing staff no longer needs to worry about

upgrading software. Instead, they will be free to

focus further on innovation.

Cloud services are applications or services

offered using cloud computing. Cloud services

delivery models include:

• Software-as-a-Service (SaaS): the cloud service

application runs on the servers of the cloud

provider. Users access the service via a Web

interface or by using an API.

• Platform-as-a-Service (PaaS): businesses

develop and deploy their business applications in

a cloud environment by using software tools

offered by the cloud provider, who is responsible

for maintaining and managing the cloud

infrastructure.

• Infrastructure-as-a-Service (IaaS): businesses

rent compute, storage, and network resources and

access them across the Internet or via a private

network.

Training organizations and academic institutions

can take advantage of cloud computing to deploy e-

Learning and mobile learning platforms that are rich

in multimedia content and with high storage

requirement without being concerned with the issues

of infrastructure management, software and

hardware upgrade. Cloud computing can boost

readiness of learning solutions by allowing teachers

and students to access the learning platform from

anywhere using internet-enabled devices. By

lowering operation costs through cloud computing, a

training organization can develop more in-house

learning content, access open learning content, or

purchase content from third parties learning material

providers.

The SaaS delivery model, as demonstrated by the

offerings of the main cloud providers and players, is

the most appropriate cloud-based solution for

implementing both e-Learning and m-Learning

platforms that include Learning Management Systems

(LMS), learning material repositories, authoring tools,

and collaboration solutions like video conferencing

and screen sharing. Indeed, educational institutions

can easily and quickly implement SaaS-based

solutions without incurring the maintenance costs,

normally inherent to in-house solutions, while

benefiting from the latest software updates and new

features offered by the cloud provider.

For mobile learning solutions to be successful, it

is important to create compelling, engaging and

connected mobile learner experiences. Therefore,

backend components or services need to feed the

mobile learner application with relevant learning

material and allow interactions between learner and

instructor, between learners, and more. With cloud

computing solution, it is becoming possible to

develop backend solutions that scale to meet the

growing demand. By leveraging the cloud

infrastructure, solutions can be implemented without

having to worry about things such as managing

machines or load balancing.

Typical requirements for a compelling mobile

learning solution include: (1) support for multiple

devices, (2) storage, retrieval, and processing learning

content outside of mobile devices, (3) integration of

new learning content from third party providers, (4)

user authentication, and (5) high scalability.

5 FRAMEWORK OVERVIEW

5.1 Context

We are considering a learning environment where

learners take courses asynchronously. In this

context, learners have personal desires to enhance

their knowledge and careers. They often have long-

term learning plans and do prefer flexible and

personalized learning environments that take into

consideration their preferences. Most of these

learners are always mobile and require having access

to their learning material from everywhere, anytime,

and using diverse mobile devices. So, the context of

this work takes into consideration the following

requirements: asynchronous learning, mobile

learners, heterogeneous mobile devices, tailored

courses, and personalized interfaces with a similar

look and feel.

5.2 System Architecture

In this context, we are proposing a cloud-based

A Cloud-based Framework for Personalized Mobile Learning Provisioning using Learning Objects Metadata Adaptation

371

framework for implementing an adaptive mobile

learning system that supports mobile learners

connecting to the learning platform using various

mobile devices. The cloud service of the training

organization has the following components: the

Profile Manager (PM), the Course Delivery

Manager (CDM), the Device Profile Manager

(DPM), and The Learning Objects Manager (LOM).

The learner’s device has the following components:

The Learner Profile Manager (LPM), the Learner

Course Delivery Manager (LCDM), and the Learner

Tutoring Interface Manager (LTIM).

PM manages the knowledge concerning the

learners. Its main tasks are:

• Performing user authentication

• Acting as a central register for registering each

new learner.

• Managing and assuring the consistency of the

databases containing learners’ profiles.

• Receiving service requests from terminals and

giving access to user profiles data.

• Initiating the LPM on the remote device.

• Checking the version of LPM and LTIM that

resides on remote devices and automatically

download any necessary updates.

DPM manages the knowledge concerning the

various devices. Its main tasks are:

• Acting as a central register, where each new

learner device must be registered.

• Managing and assuring the consistency of the

databases containing devices profiles.

CDM manages knowledge about courses and

teaching strategies. Its main tasks are:

• Providing an interface for defining learning

objects and courses knowledge (study guide and

study plan).

• Receiving service requests from devices and

giving access to courses material.

• Generating the course study guide and study plan

based on the user profile and the teaching

strategy.

• Packaging the course teaching material

according to the user profile and device profile.

• Initiating the LCDM and the LTIM on the

learner’s device.

LOM is in charge of managing in-house learning

objects and access to third parties learning objects.

LPM carries and manages a local copy of the

learner profile (preferences and learner’s model) on

the learner’s device. Its main tasks are:

• Providing the other components of the learner’s

device (LCDM and LTIM) with the learner

information (profile, identification).

• Managing and synchronizing the learner profile

information with the training organization cloud

service.

• Providing the local personalization of the

learning material. In collaboration with the

LCDM and LTIM, LPM ensures the display of

the learning material according to the learner’s

preferences and her device capabilities.

Figure 1: Architecture overview.

CSEDU 2016 - 8th International Conference on Computer Supported Education

372

LCDM manages the course delivery to the

roaming user. The main tasks of the course provider

are:

• Carrying and managing the learning material and

study plan.

• Providing a personalized learning service to the

learner based on her model and learning style.

• Synchronizing the learning content with the

educational institution cloud service.

• Ensuring the adaptation and packaging of

external learning content. Since the system is

open to third-party providers, external learning

material may need to be converted to the

required format and adapted to the learner and

device profiles.

LTIM is in charge of maintaining the profile of

the learner’s device and ensures that the display of

services is done according to the user preferences

and device capabilities.

The back-end databases of the framework

include:

• Learners’ Profile Repository: for each learner the

system maintains a profile that has two

components, the learner’s model and the learner

preferences regarding the learning style,

interfaces and content display.

• Devices’ Profiles Repository: contains for each

device a description of its features and

capabilities that are useful for the learning

service provision (screen size, bandwidth limit,

colors, resolution, etc.). Some features that can

be automatically detected by the system

(Operating System, Browser, Plug-ins) are not

stored in the repository but integrated to the

profile when initializing the Learner Tutoring

Interface Manager.

• Learning Objects Repository: contains the

learning material defined as learning objects.

• Learning Material Database: for each unit of the

learning material the system maintains its study

guide and its study plan structures.

5.3 Knowledge Structures for Adaptive

Delivery of Learning Material

Study Guide: the learning material content is

organized around a set of concepts. Each concept

has a teaching material associated with it that comes

either from an external provider or an in-house

learning object. The study guide defines the

relationships between these concepts. The

relationships consist of prerequisite, similarity and

substitute relationships.

Study Plan: The learning material structure is

organized into units and sections, and for each

section a set of concepts is learned using different

tasks: readings, labs, and tests. The study plan

defines the sequencing of the learning material

content and the time constraints and deadlines for

different tasks of the learning process.

Learner’s Model: a fuzzy overlay model based on

the learning material concepts. The model represents

static beliefs about the learner and simulates the

learner’s reasoning in some cases. With each

concept in the model is associated a fuzzy value that

represents the assessment of the learner’s knowledge

regarding this concept. Two different versions of the

learner’s model are used by the system: a global

model and a local model. The global learner’s model

is stored in the Learner Profile Repository and

represents concepts reported to the system about the

learner or learned from different learning materials.

This model represents in addition to the concepts

and associated fuzzy values, the relationships

between concepts (prerequisite, similarity, and

substitute). The local learner’s model is managed by

the user agent within the learner’s device and is

related to a specific learning material. This model

represents only the course concepts and associated

values. It is refined based on the learner’s interaction

with the system when reading the learning material

and doing the assessment exercises. The local model

is also used to update the global learner’s model.

The local model is initialized from the global model.

Teaching Strategy: A set of rules that implement

the adaptation controls for the learning material. It

consists of rules for sequencing the learning

components, rules for adding or dropping learning

components and rules for selecting between similar

or equivalent learning components.

6 ADAPTATION OF THIRD

PARTY PROVIDERS CONTENT

In traditional learning systems, learning content is

organized into several chunks called courses. Each

one may last up to one or several hours. In modern

learning systems, however, content is built from

smaller units of learning, lasting for few minutes,

called learning objects (LOs). LOs allow

customizing courses for each learner or even for a

whole organization. A single learning object may be

used in diverse contexts for multiple purposes. In

LO repositories, each LO has descriptive metadata

allowing it to be easily found in a search and

A Cloud-based Framework for Personalized Mobile Learning Provisioning using Learning Objects Metadata Adaptation

373

integrated with other LOs to build much larger units

of learning. LOs metadata can be written in XML or

any other proprietary format. The IEEE 1484.12.1

Standard for Learning Object Metadata is an

internationally recognized open standard for

describing LOs (IEEE, 2002). Pertinent attributes of

LOs include the type of object, author, owner,

format, and pedagogical attributes such as

interaction style.

The proposed framework is open to third party

providers that can provide their learning material as

learning objects. Their content can be plugged into

the framework. With the availability of open

learning objects repositories, it is now possible to

search for appropriate learning content that can be

integrated with local content based on the mobile

learner’s profile and the study plan. Examples of

such repositories are MERLOT (MERLOT, 2016)

and WISC-ONLINE (WISC-ONLINE, 2016).

Heterogeneity in LOs metadata requires a

process for adapting the learning content, built from

various in-house and imported LOs, to the learner’s

profile and device profile. The framework uses a

common metadata model to describe LOs. This

process, depicted in figure 2, involves two steps:

Step 1: metadata adapters are used to translate

imported LOs metadata models used by third parties

LO repositories to that common metadata model.

Figure 2: LOs metadata adaptation process.

Step 2: The metadata of imported LOs, described

using the common model of the framework, and the

learner’s profile and the device profile are the input

of the adaptation (transformation) service, which

generates content adapted to the learner’s device.

Open source software tools for developing mobile

applications such as PhoneGap and Apache Cordova

allow generating native apps for different kind of

platforms (Android, iOS, etc.) and various kinds of

devices. The adaptation service may use this kind of

tools for generating adapted content to the learner’s

device.

7 CONCLUSIONS

Training organizations, which offer mobile-learning,

face the following challenges: support and

adaptation of learning content to multiple devices,

adaptation of the learning material to the mobile

learner profile and preferences, and the ability to

integrate learning content from external training

providers and open repositories with in-house

learning content.

In this paper, we have described our proposed

cloud-based framework for adaptive mobile

learning. We described the components of the

framework at the cloud service of the educational

institution and the components to deploy on the

device of the mobile learner. The proposed solution

takes advantage of the benefits procured by the

cloud regarding elasticity of resources and

scalability by supporting a large number of mobile

learners. It permits to adapt both the course content

and the mobile learner interface dynamically. We

have also described the process of importing

learning objects from third party learning providers

and their integration with in-house learning objects,

which allows building and adapting the study plan

and the learning material to the mobile learner

profile.

As a future work, we intend to build a prototype

of the proposed framework and to experiment with

few LO repositories to create dynamic content.

REFERENCES

Brusilovsky, P. (2003). Adaptive navigation support in

educational hypermedia: The role of learner

knowledge level and the case for meta-adaptation,

British Journal of Educational Technology, 34(4),

pages 487–497.

CSEDU 2016 - 8th International Conference on Computer Supported Education

374

Cavus, N. and Ibrahim D. (2009). M-Learning: an

experiment in using SMS to support learning new

English language words, British Journal of

Educational Technology, 40(1), pages 78-91.

Chung, T. S., Rust, R. T., and Wedel, M. (2008). My

Mobile Music: An Adaptive Personalization System

for Digital Audio Players, Marketing science, 28(1),

pages 52–68.

Crompton, H. (2013). A historical overview of mobile

learning: Toward learner-centered education, In Z.L.

Berge & L. Y. Muilenburg (Eds.), Handbook of

mobile learning, pages 3–14, Florence, KY:

Routledge.

Davies, B. S., Rafique, J., Vincent, T. R., Fairclough, J.,

Packer, M. H., Vincent, R., and Haq, I. (2012). Mobile

Medical Education (MoMEd) - how mobile

information resources contribute to learning for

undergraduate clinical students - a mixed methods

study, BMC Med Educ, 12(1), pages 1.

De Bra, P. and Ruiter, J. P. (2001). AHA! Adaptive

Hypermedia for All, In Proc. of WebNet—World

Conference on the WWW and Internet, Association for

the Advancement of Computing in Education, pages

262–268.

Economides, A. (2006). Adaptive Mobile Learning,

Fourth IEEE International Workshop on Wireless,

Mobile and Ubiquitous Technology in Education,

WMUTE '06, pages 26–28.

Fang, H., Liu, J., Huang, R., and Li, Y. (2009). Research

on adaptive mobile learning resources platform based

on learning object model, the Int. Conference on

Application of Information and Communication

Technologies, AICT 2009, pages 1–6.

Gipple, J. and Lord, E. (2016). Understanding Mobile

Learning and Best Practices, ICS Learning Group.

http://www.icslearninggroup.com/whitepapers/underst

anding-mobile-learning-and-best-practices/.

Heinrich, P. (2012). THE IPAD AS A TOOL FOR

EDUCATION, A study of the introduction of iPads at

Longfield Academy, Kent. http://www.emergingedtec

h.com/2012/07/study-finds-benefits-in-use-of-ipad-as-

educational-tool/

Huang, J.J.S., Yang, S.J.H., Chen, Z.S.C., Wu, F.C.C.

(2012). Web content adaptation for mobile device: A

fuzzy-based approach, Knowledge Management & E-

Learning: An International Journal, 4(1), pages 102–

122.

IEEE. (2002). IEEE Standard for Learning Object

Metadata, IEEE Std 1484.12.1-2002, pages i-32, doi:

10.1109/IEEESTD.2002.94128.

Jung, H., Park, S., and Chung, K. S. (2006). An

Architecture for Adaptive Mobile Learning, In Proc.

of the 20th Int. Conf. on Advanced Information

Networking and Applications, AINA 2006, 2, pages

219–223.

Lehman, R. (2007). Learning object repositories, New

directions for adult and continuing education,

2007(113), pages 57–66.

Litchfield, B. C., Driscoll, M. P., and Dempsey. J. V.

(1990). Presentation sequence and example difficulty:

Their effect on concept and rule learning in computer-

based instruction, Journal of computer-based

instruction, 17, pages 35-40.

Masud, M. A. H. and Huang, X. (2013). M-learning

Architecture for Cloud-based Higher Education

System of Bangladesh, Mobile Computing, 2(4), pages

84–94.

MERLOT. (2016). Multimedia Educational Resource for

Learning and Online Teaching. http://www.merlot.org.

Oh, J., Herrera, J., Bryan, N. J., Dahl, L., and Wang, G.

(2010). Evolving the mobile phone orchestra, In Proc.

of the Int. Conf. on New Interfaces for Musical

Expression, pages 82–87.

Pimmer, C. and Gröhbiel, U. (2008). Mobile Learning in

corporate settings. Results from an Expert Survey,

mLearn2008, The Bridge from Text to Context.

Tennyson, R. D. and Christenson, D. L. (1988). MAIS: an

intelligent learning system, In D. H. Jonassen (Eds.),

Instructional Designs for microcomputer courseware.

Hillsdale: N. J.: Erlbaum.

Traxler, J. (2005). Defining mobile learning, IADIS

International Conference on Mobile Learning.

Vassileva, J., and Deters, R., 1998. Dynamic Courseware

Generation on the WWW, British Journal of

Educational Technologies, 29(1), pages 5-14.

Velev, D.M. (2014). Challenges and opportunities of

Cloud-based Mobile Learning, Int. J. of Information

and Education Technology, 4(1), pages 49-53.

Weber, G. & Brusilovsky, P. (2001). ELM-ART: An

adaptive versatile system for Web-based instruction,

Int. J. of Artificial Intelligence in Education, Special

Issue on Adaptive and Intelligent Web-based

Educational Systems, 12 (4), pages 351-384.

WISC-ONLINE. (2016). Wisc-Online By teachers, for

students, get unstuck. https://www.wisc-online.com/

A Cloud-based Framework for Personalized Mobile Learning Provisioning using Learning Objects Metadata Adaptation

375