OPPIA: A Context-Aware Ubiquitous Learning Platform to Exploit

Short-Lived Student Networks for Collaborative Learning

Jack Fernando Bravo-Torres

, Vladimir Espartaco Robles-Bykaev

, Martín López-Nores

Esteban Fernando Ordoñez-Morales

, Yolanda Blanco-Fernández

and Alberto Gil-Solla

Centro de Investigación e Innovación en Ingeniería, Universidad Politécnica Salesiana,

Calle Vieja 12-30 y Elia Liut, Cuenca, Ecuador

AtlantTIC Research Center, University of Vigo, Vigo, Spain

Keywords:

Sporadic Learning Networks, Ubiquitous Learning, Technology-enhanced Learning, Collaborative Learning,

Smart Learning Environments, Extended Classroom.

Abstract:

We present the OPPIA platform, a ubiquitous and smart learning environment, that deploys sporadic learning

networks (SLNs) among people (students, teachers and experts) who happen to be in a common place or in

a remote place but connected to the platform through the Internet. The idea is to establish dynamic learn-

ing networks that encourage their members to work together and create a learning environment with proper

resources and activities to satisfy their learning needs. This paper describes the design of the platform from

the lowest levels of establishing connections among members of the SLNs, up to the highest level of smart

learning services.

1 INTRODUCTION

The advance of electronic devices, wireless technolo-

gies and communication networks is changing the

way in which we interact with our environment. On

the one hand, the presence of electronic devices and

microelectronic systems embedded in everyday ob-

jects in our lives (mobile phones, laptops, cameras,

audio and video devices, clothing, vehicles, etc.) gen-

erates environments rich in sources of digitized in-

formation, accessible anywhere and anytime. On the

other hand, the omnipresence of the Internet and the

boom of the social networks create new ways of so-

cial interaction, breaking the time-space limitations of

face-to-face relationships.

This new social and technological context also

affects the learning and teaching processes. People

have unlimited access to data, information, and re-

sources through their mobile devices, carrying out

of the classroom learning and teaching opportunities,

and becoming their environment in a extended class-

room (Loureiro and Bettencourt, 2014). Moreover,

the Internet is allowing to generate learning networks

among people located hundreds of kilometers from

each other, giving support to different ways of social

and collaborative learning (Vassileva, 2008), (Memeti

and Cico, 2014).

Different approaches have been proposed in the

literature about how to generate environments that en-

courage people to improve their learning, considering

their diferences in prior knowledge, background, and

interests; and giving them access to learning contents

and collaborative learning networks at the right time,

at the right place, and in the right form (Paraskakis,

2005), (Al-Zoube, 2009), (Spector, 2014). For in-

stance, many universities are using Learning Man-

agement Systems (LMS) (Coates et al., 2005), (Mah-

negar, 2012) and Distributed Learning Environments

(DLE) (Alavi, 2004) as a way to give to students

and teachers a wide access to different resources

to increase the performance of their formal learn-

ing processes. On the other hand, Personal Learn-

ing Environments (PLE) (Humanante-Ramos et al.,

2015), (Dabbagh and Kitsantas, 2012) are being de-

signed to allow people to create their own learning en-

vironments to satisfy their interests —most typically,

in informal learning scenarios. Similarly, several pro-

posals take into account the learning behaviour of

the students and the context to recommend differ-

ent resources according with their needs and pro-

ﬁles (Mengmeng Li, 2013).

This paper is about applying technologies to

enable new forms of learning interactions outside

the classroom. Speciﬁcally, we are building a plat-

494

Bravo-Torres, J., Robles-Bykaev, V., López-Nores, M., Ordoñez-Morales, E., Blanco-Fernández, Y. and Gil-Solla, A.

OPPIA: A Context-Aware Ubiquitous Learning Platform to Exploit Short-Lived Student Networks for Collaborative Learning.

In Proceedings of the 8th International Conference on Computer Supported Education (CSEDU 2016) - Volume 1, pages 494-498

ISBN: 978-989-758-179-3

form called OPPIA ("OPPortunistic Intelligent Ambi-

ent learning") that aims at facilitating the creation and

exploitation of Sporadic Learning Networks (SLN),

communicating each individual with the people (stu-

dents, teachers, or experts) that surround him/her or

connected through the Internet at a given moment,

and considering the information and resources (both

institutional and personal) that may be relevant to

them, based on their proﬁles, formal curriculum, ex-

pertise, and learning interests. The goal is to provide

an ubiquitousand smart learning environment that en-

courages students to collaborate in their learning pro-

cess, sharing knowledge and resources to establish

dynamic learning networks that are born and die ac-

cording to their learning needs.

OPPIA harnesses advances in knowledge-based

recommender systems and expert systems (i) to es-

tablish learning networks among those people that

can help overcome the learning needs identiﬁed by

the platform or triggered by some student, and (ii)

to provide tailored made learning services driven by

the user’s context, by students/teacher/expert proﬁles,

by the institutional learning information (learning ob-

jects, general curriculum, teaching scheduling,...), by

the personal resources of the students, and by any

information about their learning interests. OPPIA

allows the connected devices to share resources (e.g.

networking capabilities, storage space, computer and

processing capabilities, simulation software,...) in

order to generate virtual and distributed laborato-

ries, improving the learning opportunities of the SLN

members and overcoming speciﬁc limitations of their

handheld devices.

In addition, our platform uses data mining tech-

niques, personalization capabilities and Semantic

Web technologies, in order to efﬁciently manage the

contents available to SLNs and customize virtual

learning environments of each of its members. Con-

tent and learning activities are offered to each member

of the network in the format that best ﬁts their learn-

ing style and considering the characteristics of their

mobile devices. New contents are generated dyna-

mically, integrating existing institutional and personal

resources to satisfy the personal learning needs of the

students.

The paper is organized as follows. In Section II,

we present the architecture of the OPPIA platform,

putting the focus on the functionality of the mecha-

nisms hosted in each level. Next, Section III describes

one application scenario of OPPIA to illustrate how

our SLNs improve the learning process among stu-

dents in a university environment. Finally, Section IV

concludes the paper and points out some lines of fur-

ther work.

2 THE OPPIA PLATFORM

The OPPIA platform relies on an fully interactive

multi-layer architecture, organized in several layers

and services. Conceptually, its architecture has ﬁve

levels (see Fig. 1) that will be described in the follow-

ing subsections.

Figure 1: The conceptual layers of the OPPIA platform.

2.1 Communication Layer

This layer is responsible for providing the necessary

mechanisms to establish connections proactively and

transparently to the users whenever deemed appropri-

ate by the information from higher levels of the ar-

chitecture. So, the sporadic learning networks rely

ﬁrstly on ad-hoc networks laid dynamically among

the mobile devices of the people (students, teacher or

experts) who happen to be close to one another at a

given moment. Communications with network mem-

bers in remote locations, with knowledge bases and

learning services (upper layers of our platform) are

achived through links to hotspots or 3G/4G connec-

tions available to any members of the network. All

protocols and mechanisms for establishing links and

maintaining the necessary QoS levels are housed in

this layer, too.

2.2 Cloud Computing Layer

The second layer aims at enabling efﬁcient sharing of

the resources available to each device within an SLN

or located in areas close to them and participating in

OPPIA: A Context-Aware Ubiquitous Learning Platform to Exploit Short-Lived Student Networks for Collaborative Learning

495

other SLNs. The tandem between mobile devices and

cloud computing works perfectly because handheld

terminals are constrained by their processing power,

battery life and storage capabilities, whereas cloud

computing can provide the illusion of unlimited com-

puting resources (Mell and Grance, 2011), (Kim et al.,

2011).

The extra resources required by the handheld de-

vices can be provided either by (i) centralized servers

in the cloud, depending on connectivity to the Inter-

net, or (ii) cloudlets supported by ﬁxed nodes at the

edge of the Internet or by capable mobile terminals

connected in the ad-hoc network. With this in mind,

we take advantage of the concept of Sporadic Cloud

Computing (SCC), presented in (OrdoÃ

sez-Morales

et al., 2015), (Ordonez-Morales et al., 2015), in which

the user’s devices exploit both the resources avail-

able in the remaining terminals connected to the ad-

hoc network (computing, storing, networking, sens-

ing,...), and those provided from external data cen-

ters. In our platform, SCC allows to generate vir-

tual and distributed laboratories conformed with ex-

isting resources in the member’s devices of the dif-

ferent SLNs, who are physically close to each other.

This avoids — as far as possible — dependence on

access to the Internet to perform tasks and gives stu-

dents access to specialized software not suitable for

low performance devices. So, our cloud computing

layer provides the following services:

• Storing information in spaces in the cloud, linked

to source/target devices, creating/consuming

users, location, etc.

• Accessing and serving information of high-level

user proﬁles during the formation of ad-hoc net-

works.

• Synchronizing multiple ﬂows of information

coming from the connected devices.

• Managing of the simulation and programming re-

sources available on the users’ mobile devices so

they can be used in a transparent manner by SLN

members (virtual and distributed laboratories).

• Providing access to cloud services on the Internet:

databases, semantic repositories, physical labora-

tories, etc.

2.3 Knowledge Management Layer

Our platform uses information derived from personal

or institutional sources to provide users with the best

resources (according to their personal learning styles

and characteristics of their access devices) and activ-

ities (both individual and group) that stimulate their

learning and allow them to increase their academic

achievement or satisfy their learning needs. The

"Knowledge Management" layer is the place to put

solutions from the areas of data mining, recommender

systems and the Semantic Web to automatically select

the best proﬁles to form the learning network, choose

the pieces of information for the greatest beneﬁt of

the members of the SLN, while personalizing the con-

tents delivered by each device and the activities to be

performed by the group. To do this, it is necessary

to rely on techniques for modelling the user’s prefer-

ences, considering different proﬁles (students, teach-

ers, experts, personal devices...) and contents (insti-

tutional and personal). Moreover, in this modelling

process, OPPIA takes advantage of the academic in-

formation stored in the institutional databases such

as general curriculum (containing several academic

guidelines based on career, skills, course,...), moni-

toring teaching activities, and learning outcomes.

In OPPIA, the contents are modeled through Dy-

namic Reusable Learning Objects (DRLOs) (Valder-

rama et al., 2005), provided by the institution (insti-

tutional DRLOs) or from students, teachers or the In-

ternet. In the same way, we need to use recommen-

dation strategies that select the most appropriate con-

tents for each member or group of members of our

SLNs. In addition, we need modelling techniques to

infer knowledge about the future learning interests of

the SLN members by keeping track their academic ac-

tivities, learner’s web surﬁng habits and preferences,

and proﬁles in traditional social networks (obviously,

with the explicit consent from the users). Finally, for

the efﬁcient management of the metadata associated

with the learning process, information storage, analy-

sis and inferences, we need to use learning ontologies,

especially designed for this purpose.

2.4 Expert Systems Layer

To achieve the desired results, both in motivation and

performance of the member of a sporadic learning

network, OPPIA relays selection and design of con-

tents, educational resources, and learning activities to

the "Expert Systems" layer. With this aim, the expert

system incorporates an assembler able to create DR-

LOs. The educational institutions (universities, col-

leges, institutes, schools,...) create ofﬁcial DRLOs

— developed in different formats (video, image, text,

audio,...) to meet the learning styles of students —

that cover the main contents related to the curriculum.

In turn, the DRLO repository can be expanded with

learning objects from users themselves or obtained

from the Internet. Furthermore, OPPIA has the ability

to produce new learning objects and educational re-

sources, from DRLOs existing in the repository. For

CSEDU 2016 - 8th International Conference on Computer Supported Education

496

this, the proﬁle Analyzer, the Resource Analyzer and

the Assessment Analyzer are used.

2.5 Learning Services Layer

In order to access the different services and learning

objects that are provided by our intelligent learning

ambient, the services layer incorporates three main el-

ements: a set of intelligent ICTs (mobile apps, desk-

top and web applications, including functionalities as

sporadic chats, forums, renderers of learning objects,

etc.), adaptable classrooms (virtual classrooms that

are set to the user proﬁles), and smart learning plans

that are dynamically designed for a particular learning

session.

3 SAMPLE APPLICATION

SCENARIO

In this section, we describe a speciﬁc sample scenario

where a set of university students collaborate to im-

prove their performance in maths. Events occur in the

reading room of the central library of the university.

As usual, manystudents are with their laptops, debug-

ging and understanding class notes, developing their

tasks or preparing their exams. John, a ﬁrst-year stu-

dent of electronics engineering, is solving some prob-

lems of differential calculus. The results of his recent

tests showed that he needs a lot of work if he wants to

pass the course. Like John, other students from differ-

ent groups are working on the same subject; however,

because they are located in different parts of the read-

ing room and belong to different classes, they ignore

the situation, making it impossible for them to work

together to improve their academic performance. In

this situation, OPPIA can manage the creation of a

sporadic learning network among students with sim-

ilar weaknesses, organizing their work and providing

resources in a way that they will be systematically ac-

cessing an increasingly more complex knowledge.

To this end, ﬁrstly OPPIA accesses the knowl-

edge base to analyze John’s proﬁle. Speciﬁcally, it

seeks to determine what kind of learning resources

(video, image, voice, text, ...) John has used more of-

ten. Likewise, it checks the results of his latest test

of differential calculus, discovering that John’s main

shortcomings lie in his prior knowledge of analytic

geometry, rather than understanding the concepts of

calculus (based on feedback from his teacher). With a

similar procedure, OPPIA analyzes the proﬁle of the

other students present in the library. Through the Ex-

pert Systems layer, our platform selects students who

foresee a better learning outcome and send them an

invitation to form the SLN. The message contains the

session length and subject matter. With the conﬁrma-

tion of the members, the SLN is formed, proceeding

to unfold the virtual workplace.

When John accesses his virtual workspace, he no-

tices that the platform has assigned a work plan di-

vided into three parts: individual activities, group ac-

tivities and tutorial session. In OPPIA, the "Learning

Services" layer generates custom work plans for each

SLN member, coordinating and synchronizing the

group and personal activities. In the case of John, his

personal activities include reviewing several videos

explaining the theme, the study of multiple slides with

a summary of the main concept, several mathematical

games and the development of multiple-choice tests.

As can be seen, OPPIA has chosen these activities and

resources based on John’s visual learning style.

Upon completion of individual activities, the SLN

members move on to group activities. Students are

challenged to solve a series of problems aimed at

putting into practice the concepts learned in their

individual work. To do this, John and the other

SLN members have communication tools (chat, group

video conferences, forums) that allow them to interact

and address each of the questions that the system will

propose to them.

Finally, OPPIA programs a virtual tutoring with

one of the professors of the Department of Mathemat-

ics, who is connected to the platform at that moment.

Because the platform has access to teacher hours, the

system sends a request message to schedule a consul-

tation through a virtual classroom with students of the

SLN. This virtual class, reinforces the knowledge de-

veloped during the working session of the students.

After the virtual class, the session is terminated and

the learning network is dissolved.

4 CONCLUSIONS

The OPPIA platform aims to provide the necessary

mechanisms to exploit the potential of short-lived

learning networks to improve the academic perfor-

mance of students and satisfy their learning needs.

Our platform greatly enhances the learning experi-

ence of the SLN members through (i) an appropri-

ate selection of those proﬁles that will provide greater

support to the development of the learning tasks, (ii)

the sharing of the resources available on students de-

vices that allow to create virtual laboratories that may

be used transparently by users of the platform, and

(iii) the design of resources and learning activities

based on the analysis of the student proﬁles, academic

performance, study schedules, and personal learning

OPPIA: A Context-Aware Ubiquitous Learning Platform to Exploit Short-Lived Student Networks for Collaborative Learning

497

interests of the SLN members.

ACKNOWLEDGEMENTS

The authors from Universidad Politécnica Salesiana

have been supported by the "Sporadic Networks

to provide Information Services for Next Genera-

tion Users on Motion" and "Sistemas Inteligentes de

Soporte a la Educación" research projects (CIDII-

010213). The authors from the University of Vigo

have been supported by the European Regional De-

velopment Fund (ERDF) and the Galician Regional

Government under agreement for funding the Atlantic

Research Center for Information and Communication

Technologies (AtlantTIC), as well as by the Ministe-

rio de EducaciÃ¸sn y Ciencia (Gobierno de EspaÃ

sa)

research project TIN2013-42774-R (partly ﬁnanced

with FEDER funds).

REFERENCES

Al-Zoube, M. (2009). E-learning on the cloud. Int. Arab J.

e-Technol., 1(2):58–64.

Alavi, M. (2004). Distributed learning environments. Com-

puter, 37(1):121–122.

Coates, H., James, R., and Baldwin, G. (2005). A criti-

cal examination of the effects of learning management

systems on university teaching and learning. Tertiary

education and management, 11:19–36.

Dabbagh, N. and Kitsantas, A. (2012). Personal learning

environments, social media, and self-regulated learn-

ing: A natural formula for connecting formal and in-

formal learning. The Internet and Higher Education,

15(1):3 – 8. Social Media in Higher Education.

Humanante-Ramos, P., Garcia-Penalvo, F., and Conde-

Gonzalez, M. (2015). Personal learning environments

and online classrooms: An experience with university

students. Tecnologias del Aprendizaje, IEEE Revista

Iberoamericana de, 10(1):26–32.

Kim, S., Song, S.-M., and Yoon, Y.-I. (2011). Smart learn-

ing services based on smart cloud computing. Sensors,

11(8):7835.

Loureiro, A. and Bettencourt, T. (2014). The use of vir-

tual environments as an extended classroom: A case

study with adult learners in tertiary education. Proce-

dia Technology, 13:97 – 106. {SLACTIONS} 2013:

Research conference on virtual worlds. Learning with

simulations.

Mahnegar, F. (2012). Learning management system. In-

ternational Journal of Business and Social Science,

3(12):144–150.

Mell, P. and Grance, T. (2011). The nist deﬁnition of cloud

computing.

Memeti, A. and Cico, B. (2014). Supporting content and

learner collaboration and interaction through cloud

computing models. In Computational Intelligence,

Communication Systems and Networks (CICSyN),

2014 Sixth International Conference on, pages 145–

148.

Mengmeng Li, Hiroaki Ogata, B. H. N. U. K. M. (2013).

Context-aware and personalization method in ubiqui-

tous learning system. Journal of Educational Technol-

ogy & Society, 16(3):362–373.

OrdoÃ

sez-Morales, E., Blanco-FernÃ ˛andez, Y., LÃ¸spez-

Nores, M., Bravo-Torres, J. F., Pazos-Arias, J. J., and

Ramos-Cabrer, M. (2015). Sporangium: Exploiting a

virtualization layer to support the concept of sporadic

cloud computing with users on the move. In Rocha,

A., Correia, A. M., Costanzo, S., and Reis, L. P., ed-

itors, New Contributions in Information Systems and

Technologies, volume 353 of Advances in Intelligent

Systems and Computing, pages 959–966. Springer In-

ternational Publishing.

Ordonez-Morales, E., Bravo-Torres, J., Saians-Vazquez, J.,

Blanco-Fernandez, Y., Lopez-Nores, M., and Pazos-

Arias, J. (2015). Sporangium - validating the con-

cept of sporadic social networks in pervasive applica-

tions. In EUROCON 2015 - International Conference

on Computer as a Tool (EUROCON), IEEE, pages 1–

Paraskakis, I. (2005). Ambient learning: a new paradigm

for e-learning. In The 3rd International Conference on

Multimedia and Information & Communication Tech-

nologies in Education (m-ICTE2005), Spain, pages

26–30.

Spector, J. (2014). Conceptualizing the emerging ﬁeld of

smart learning environments. Smart Learning Envi-

ronments, 1(1):2.

Valderrama, R. P., Ocaña, L. B., and Sheremetov, L. B.

(2005). Development of intelligent reusable learning

objects for web-based education systems. Expert Sys-

tems with Applications, 28(2):273–283.

Vassileva, J. (2008). Toward social learning environ-

ments. Learning Technologies, IEEE Transactions on,

1(4):199–214.

CSEDU 2016 - 8th International Conference on Computer Supported Education

498