A Survey of Context-awareness in Learning Environments in 2010-2016

Aziz Hasanov and Teemu H. Laine

Department of Software, Ajou University, Suwon, Republic of Korea

Keywords:

Context-aware, Education, Learning Environment, Survey.

Abstract:

Context-aware learning environments can detect the learner’s context and adapt learning materials to match the

context. The support for context-awareness is essential in these systems so that they can make learning con-

textually relevant. Previously, several surveys on context-aware learning environments have been conducted,

but they are either old or they do not consider several important aspects of context-awareness. To alleviate this,

we ﬁrst performed a literature search on context-aware learning environments in 2010-2016. After ﬁltering

the results, we analyzed 28 studies. Highlights of the results are: (i) PDAs and mobile phones are the most

common client types, (ii) RFID/NFC are the most common sensors, (iii) ontology is the most common context

modeling approach, and (iv) context data typically originates from the learner’s proﬁle or the learner’s loca-

tion. Additionally, we proposed a taxonomy for context categories in context-aware learning environments.

Finally, based on our survey results, we gave directions for future research in the ﬁeld. These results can be of

interest to educational technology researchers and context-aware application developers.

1 INTRODUCTION

Recent development of advanced information tech-

nologies, such as wireless communications, sensors,

and the Internet of Things, has enabled adaptive

learning, including mobile and context-aware learn-

ing. In this paper, we use the deﬁnition of con-

text as a set of entities that constitute the learner’s

situation (Laine and Nygren, 2016). Examples of

these contextual entities in a learning environment

are current place, time, other nearby learners, learn-

ing style, and learning history. The abilities of a

learning environment to detect the learner’s context

and to adapt its behavior accordingly play a crucial

role in personalized learning (G

omez and Fabregat,

2010). Context-awareness can make a signiﬁcant dif-

ference in learning efﬁciency compared to traditional

classroom-based learning because in context-aware

learning environments learning resources and activi-

ties are adapted to match the learner’s current situa-

tion. A body of studies in the ﬁeld of context-aware

learning show positive effects of context-aware tech-

nologies on learning (Hsu et al., 2016; Gomez et al.,

2016; Wu et al., 2012).

Several surveys have been conducted on context-

aware systems with diverse perspectives. A litera-

ture review on mobile and ubiquitous learning envi-

ronments published in 2001-2010 focused on analyz-

ing non-technical aspects, such as publication statis-

tics, target groups, and learning domains (Hwang and

Tsai, 2011). Pedagogical and technical aspects of

context-aware learning environments were analyzed

in another survey (Laine and Joy, 2009). Verbert

et al. (Verbert et al., 2012) compared context-aware

recommender systems for educational purposes. Fi-

nally, technological foundations of context-aware sys-

tems have been surveyed from the perspectives of

mobile learning architectures (Baccari et al., 2015)

and context-aware middleware architectures (Li et al.,

2015). Despite these valuable surveys, there has not

been a recent study which would thoroughly analyze

contemporary learning environments from the per-

spective of context-awareness.

To understand the landscape and technical ap-

proaches of context-aware learning environments

published between 2010 and 2016, we analyze 28

articles on context-aware learning environments for

different purposes. We particularly focus on identi-

fying and comparing the technical solutions through

which context-awareness has been established. More-

over, we use the discovered information to discuss

directions for future context-aware learning environ-

ments. Research contributions of this review are four-

fold: (i) explore and compare context-aware learning

environments in 2010-2016; (ii) identify various ap-

proaches in which context-awareness has been estab-

234

Hasanov, A. and Laine, T.

A Survey of Context-awareness in Learning Environments in 2010-2016.

DOI: 10.5220/0006255302340241

In Proceedings of the 9th International Conference on Computer Supported Education (CSEDU 2017) - Volume 1, pages 234-241

ISBN: 978-989-758-239-4

lished in these systems; (iii) propose a taxonomy of

context categories to help comparing context-aware

learning environments; and (iv) give directions for fu-

ture context-aware learning environments.

2 METHODOLOGY

In this survey we employed a simpliﬁed version of the

systematic literature review methodology (Kitchen-

ham and Charters, 2007). The original methodology

comprises a set of guidelines for planning, conduct-

ing, and reporting a systematic literature review with

a focus on software engineering ﬁeld. In the follow-

ing, we describe our adaptation of this methodology

in terms of data collection and analysis.

2.1 Data Collection

First, we deﬁned literature search criteria. The ﬁrst-

order criteria were deﬁned as follows: a combination

of predeﬁned keywords (e.g., context-aware, educa-

tion, system); publication time range 2010-2016; type

of publication: conference and journal publications;

digital libraries / search tools: Google Scholar, IEEE,

and ACM. We also deﬁned a second-order criterion

as relevance to the ﬁeld of context-aware education,

thus selected publications should, in sufﬁcient detail,

describe a learning environment that is context-aware.

Data collection and ﬁltering were performed in

four steps using the aforementioned criteria. The ini-

tial ﬁltering of search results was based on the ﬁrst-

order criteria. In this ﬁrst step, we discovered 53 po-

tential articles. Next step of ﬁltering was based on

the second-order criterion, which was applied to titles

and abstracts of search results. Then, the same crite-

rion was applied during skimming through the previ-

ously ﬁltered results. Finally, we read thoroughly the

remaining papers and applied the second order ﬁlter-

ing criterion. As the result, we had 28 papers to be

analyzed for this survey.

2.2 Data Analysis

To analyze the ﬁndings in a structured manner, we

ﬁrst established taxonomies through which the re-

viewed systems were to be compared. These tax-

onomies would enable us to classify various aspects

of the reviewed learning environments and thereby

understand their similarities and differences. Some of

the taxonomies were discovered from previous publi-

cations and some of them we created out of necessity.

After establishing the taxonomies, we performed an

in-depth analysis of the selected papers to assign ap-

propriate values to each aspect. Finally, based on this

classiﬁcation of learning environments into different

aspects, we gave our own interpretations of the ﬁnd-

ings to provide useful ideas for future research in the

ﬁeld.

3 RESULTS

Here we present an overview of the reviewed learning

environments and describe the technical approaches

that were used in these systems to establish context-

awareness. It must be noted that despite our best ef-

forts we were not able to extract all required data from

the analyzed articles due to lack of details in their

presentation. Such cases are marked with ’n/a’ as in

not available. Non-existent aspects are marked with a

dash (’-’).

3.1 Overview

Table 1 gives an overview of the surveyed learning en-

vironments with a name (if available) and reference,

description, and client type. As the table indicates,

context-aware learning environments have been de-

veloped for various, mostly informal, learning sce-

narios and subjects to be used by both children and

adults.

By ’client type’ we refer to a device or software

through which learners use the learning environment.

We categorized clients used in the reviewed learn-

ing environments into six types: (i) Personal Digital

Assistants (PDA), (ii) Mobile phones (including also

smartphones), (iii) Tablets, (iv) Wearables, (v) Lap-

tops/PCs, and (vi) Web browsers.

Mobile devices were often used as clients in the

reviewed systems. In particular, mobile/smart phones

and PDAs were the most common clients. There

were seven systems with web browser clients, thus

making them platform independent. Only one article

proposed the use of wearables (smartwatches) in the

learning process (Santos et al., 2015).

3.2 Context-awareness

To compare the technical approaches through which

context-awareness has been established in contempo-

rary context-aware learning environments, we deﬁned

a classiﬁcation scheme using the following aspects:

Context Acquisition, Context Modeling, Context En-

tities, and Sensors. In the following, before present-

ing the results, we explain these aspects with their re-

spective taxonomies.

A Survey of Context-awareness in Learning Environments in 2010-2016

235

Table 1: Overview of context-aware learning environments.

System Description Client type

ALESS (Hsu et al., 2016) Supports active learning in a museum for elementary school students PDA

WoBaLearn (Zhang et al.,

2016)

Guides professionals in ofﬁce and factory environments to engage in

work-based learning activities

Tablet

AICARP (Santos et al.,

2015)

Provides interactive recommendations to support language learning

Wearables,

Laptop/PC

(Gomez et al., 2016)

Delivers contextualized content to the students in nursery, medicine and

systems engineering

Tablet, Mobile

phone

CAALS (Chen and Lin,

2016)

Supports active learning in museum for elementary school students Tablet

MobiSWAP (Harchay et al.,

2015)

Semantic web-based system that supports personalized self-assessment

in mobile environments for computer science students

PDA, Lap-

top/PC, Mobile

phone

(Benlamri and Zhang,

2014)

Provides knowledge-driven recommendations to learn C++ program-

ming and photography

Mobile phone

(Kim and Lee, 2014)

Provides learners with English conversation learning contents for the

business sector. Recognizes trade names from signboard images

Mobile phone

UoLmP (G

omez et al.,

2014)

Supports semi-automatic adaptation of learning activities, particularly

for learning English

Tablet, Mobile

phone

E-SoRS (Akbari and Taghi-

yareh, 2014)

Provides adapted exercises to a graduate-level students based on their

learning styles

Web browser

(Yin et al., 2013) Offers technicians learning opportunities during maintenance work

PDA, Mobile

phone

SCROLL (Li et al., 2013)

Helps Japanese language learners to record their learning logs and gives

them recommendations later

Tablet

(Kasaki et al., 2012)

A location-aware language learning with adaptive correlation comput-

ing methods

Web browser

MLAS (Chorﬁ et al., 2012)

Applies Case-Based Reasoning approach to determine appropriate con-

tent for the learner

PDA, Tablet,

Laptop/PC,

Mobile phone

SRL@Work (Siadaty et al.,

2012)

Learning environment for workers at a car manufacturer, SMEs and at

teachers’ professional association

Web browser

CAULS (Chen and Huang,

2012)

Learning in museum with elementary school teachers and students PDA

(Wu et al., 2012) Supports cognitive apprenticeships in nursing skills training PDA

(Alharbi et al., 2012) Provides a student-centric approach to lifelong learning Web browser

ePH (Vladoiu and Constan-

tinescu, 2011)

Multi-agent system that provides support for various learning scenarios

PDA, Lap-

top/PC, Mobile

phone

IWT (Capuano et al., 2011) Provides personalized e-learning Web browser

(Jia et al., 2011)

Workplace e-learning system using Key Performance Indicator and

ontology-based approaches

Web browser

(Wang and Wu, 2011)

Ubiquitous learning system that gives courseware recommendations in

a museum

PDA

(Yaghmaie and Bahreinine-

jad, 2011)

Adaptive learning system using multi-agents that adapts course topics

according to learners’ experiences

n/a

(Wang and Wang, 2011) Ubiquitous learning system based on a service-oriented architecture

PDA, Mobile

phone (with

RFID)

(Scott and Benlamri, 2010) Collaborative learning space applied to university lectures Web browser

(Yu et al., 2010) Semantic learning space infrastructure and English learning assistant Tablet

TANGO (Ogata et al.,

2010)

Supports language learning (English, Japanese, Chinese and Spanish) PDA

PCULS (Chen and Li,

2010)

English vocabulary learning based on the learner’s location, learning

time, English vocabulary abilities and leisure time

PDA

CSEDU 2017 - 9th International Conference on Computer Supported Education

236

Context Acquisition is the process of capturing the

learner’s current context and its methods vary sig-

niﬁcantly depending on available technology and the

system’s intended use of context data. There are

three fundamental ways for context acquisition (Per-

era et al., 2014): (i) context can be sensed directly

through sensors; (ii) context can be derived from

sensed raw data (or other data source) through com-

putation; or (iii) context can be provided by user in-

put. Accordingly, our context acquisition taxonomy

consists of classes Sensor, Derived and User input.

Context Modeling deﬁnes a way of representing

the context in format that can be understood and

processed by the computer. There are six context

modeling approaches that have been classiﬁed previ-

ously (Strang and Linnhoff-Popien, 2004) and used in

context-aware systems: Key-value, Markup scheme,

Graphical, Object-based, Logic-based, and Ontology-

based. In addition to this list, we added Database-

based models that employ a database (e.g., relational,

NoSQL) to store the learner’s context data. This

amendment was required because we found that sev-

eral context-aware learning environments store con-

text data in a database instead of using a speciﬁc con-

text modeling technique.

There are many ways to categorize Context Enti-

ties into taxonomies (Verbert et al., 2012; Li et al.,

2015; Constantino Martins et al., 2008; G

omez et al.,

2014) but none of them were found to be sufﬁcient for

our survey. Based on the analysis of previous work,

we established a taxonomy that comprises ﬁve con-

text entity groups: User, Technical, Spatio-Temporal,

Pedagogical and Environmental. Figure 1 illustrates

our taxonomy with example entities for each group.

Figure 1: A taxonomy of context entities.

Sensors refer to hardware and software (vir-

tual) sensors that were used in the reviewed learn-

ing environments. We identiﬁed six groups of sen-

sors that have been used as sources for context

data: RFID/NFC, GPS, Camera, Microphone, IR

(infrared)-based sensors, and network (e.g., band-

width). One of the surveyed studies (Santos et al.,

2015) only speciﬁed groups of sensors, thus we re-

ported this case as “physiological and inertial sen-

sors”.

Table 2 compares the surveyed context-aware

learning environments according to the aforemen-

tioned classiﬁcation aspects. Due to lack of technical

details available, we could not ﬁnd information for all

cells in the table.

Sensors (S) and user input (UIn) were the most

common sources for context acquisition with 20 and

19 instances, respectively, and many learning environ-

ments used a combination of the two. A typical exam-

ple would be a learning environment that asks the user

to insert learning preferences and background infor-

mation before learning, and during the learning pro-

cess the learner’s location is sensed with RFID (San-

tos et al., 2015; Zhang et al., 2016). Using these con-

text entities, the system can then provide personalized

materials that suit the learner’s preferences and loca-

tion. Derived data (D), which is based on reﬁning raw

data into information with a higher degree of abstrac-

tion, was much less common. An example of how

derived data is used is the distance between learners

or the learner and learning target, which are computed

based on location coordinates (Hsu et al., 2016).

We hypothesized that ontologies would be the

most common context modeling approach today, but

there also would be representatives of other ap-

proaches. Our analysis proved the ﬁrst part of the

hypothesis correct. Ontologies were found in 13 sur-

veyed systems, thus making them the most popu-

lar context modeling approach. Databases were also

common with 8 instances, but other approaches were

nearly non-existent. Some articles described also the

types of ontologies, such as learner ontology, device

ontology, domain ontology, and content ontology.

A myriad of context entities were employed to es-

tablish context-awareness in the reviewed learning en-

vironments. The results in Table 2 suggest that user

(U) context and spatio-temporal (ST) context were the

most common entity groups with 21 and 19 occur-

rences, respectively. A typical context entity was user

proﬁle with information such as previously studied

learning materials and learning preferences. Within

the spatio-temporal context, location of the learner

was the most common entity. Location was often used

with timestamp to determine where the learner is at a

A Survey of Context-awareness in Learning Environments in 2010-2016

237

Table 2: Context-awareness in learning environments.

System

Context

Acquisition

Context Modeling Context Entities Sensors

ALESS (Hsu et al., 2016) (S) DB (relational) (P), (ST) RFID

WoBaLearn (Zhang et al.,

2016)

(S), (UIn)

Ontology based, DB

(relational)

(U), (ST) n/a

AICARP (Santos et al.,

2015)

(S), (UIn) n/a (U), (E)

Physiological and iner-

tial sensors

(Gomez et al., 2016) (S) Ontology based (U), (P), (ST) RFID/NFC, GPS

CAALS (Chen and Lin,

2016)

(S) DB (relational) (ST) RFID

MobiSWAP (Harchay et al.,

2015)

(UIn)

Ontology based (object,

learner, context, portfo-

lio, domain)

(T), (P), (ST) -

(Benlamri and Zhang,

2014)

(S), (UIn)

Ontology based (do-

main, activity, learner,

device, environment)

(U), (T), (P),

(ST)

Network (connection/

bandwidth)

(Kim and Lee, 2014) (S) DB (relational) (ST) GPS

UoLmP (G

omez et al.,

2014)

(S), (UIn) n/a (U), (P), (ST) GPS

E-SoRS (Akbari and Taghi-

yareh, 2014)

(UIn)

Ontology based

(learner, content)

(P) -

(Yin et al., 2013)

(S), (D),

(UIn)

Markup scheme (XML)

(U), (T), (E),

(ST)

GPS, Camera

SCROLL (Li et al., 2013)

(S), (D),

(UIn)

DB (relational) (U), (T), (ST) RFID, GPS, Camera

(Kasaki et al., 2012) (S), (D) n/a (U), (ST) GPS

MLAS (Chorﬁ et al., 2012) (UIn) Markup scheme (U), (T) -

SRL@Work (Siadaty et al.,

2012)

(UIn) Ontology based (U) -

CAULS (Chen and Huang,

2012)

(S), (UIn) n/a (U), (ST) RFID

(Wu et al., 2012) (S), (UIn) DB (U), (ST) RFID

(Alharbi et al., 2012) (D), (UIn) DB (U) -

ePH (Vladoiu and Constan-

tinescu, 2011)

(S), (D),

(UIn)

Ontology based

(U), (T), (E), (P),

(ST)

GPS

IWT (Capuano et al., 2011) (UIn) Ontology based (U), (T), (P) -

(Jia et al., 2011) (UIn) Ontology based (U) -

(Wang and Wu, 2011) (S), (UIn) DB (U) RFID

(Yaghmaie and Bahreinine-

jad, 2011)

(UIn) Ontology based (U) -

(Wang and Wang, 2011) (S)

Ontology based (learn-

ing method, learning

domain, location con-

text)

(P), (ST) RFID

(Scott and Benlamri, 2010)

(S), (D),

(UIn)

Ontology based (U), (T), (ST) IR-based

(Yu et al., 2010) (S)

Ontology based (con-

text, content, domain)

(U), (T), (P),

(ST)

RFID, GPS, Camera,

Microphone

TANGO (Ogata et al.,

2010)

(S) n/a (ST) RFID

PCULS (Chen and Li,

2010)

(S) DB (relational) (U), (ST)

Network (WLAN posi-

tioning)

CSEDU 2017 - 9th International Conference on Computer Supported Education

238

given time. Technical (T), pedagogical (P) and envi-

ronmental (E) contexts were utilized in 9, 10 and 3

cases, respectively.

Location-awareness was clearly visible in pop-

ularity of sensor technologies with RFID/NFC and

GPS occurring 10 and 8 times, respectively. A typical

example of using RFID/NFC in the reviewed systems

was to provide location-sensitive learning content

when the learner reads a tag with a mobile device (Wu

et al., 2012). Adaptations of media type and size

of resources were done by sensing through network

properties such as connection type (IEEE 802.11 or

GPRS) and bandwidth (Benlamri and Zhang, 2014).

Camera and microphone were used only in a hand-

ful of systems, for example when the learner captures

his learning log or the system senses audio signals (Li

et al., 2013; Yu et al., 2010). Infrared was used for

indoor positioning (Scott and Benlamri, 2010). Some

of the reviewed systems did not use sensors or details

about sensors were omitted.

3.3 Discussion

We can infer from the results presented in Table 2 that

the most popular clients for context-aware learning

environments are PDAs and mobile phones. This is

probably due to their high mobility, which allows the

learner to traverse within a context and between con-

texts. Mobility is a great affordance for learning envi-

ronments that are based on informal learning contexts,

such as museums, science centers and parks. Techno-

logical advances of mobile devices have made them

truly smart in terms of processing power and sens-

ing capabilities. Moreover, many PDA devices and

smartphones provide RFID or NFC reader modules,

which allow learners to interact with surrounding ob-

jects. Finally, the popularity of PDAs is somewhat

surprising given the fact that smartphones have taken

over the markets of high-performance mobile devices

since the launch of the ﬁrst iPhone in 2007.

The popularity of mobile phones and PDAs was

expected given their pervasiveness in our lives. We

were surprised that the recent boom in wearable tech-

nologies was not evident in the reviewed learning en-

vironments. Wearable technologies have been iden-

tiﬁed to possess considerable affordances for learn-

ing applications (Bower and Sturman, 2015). Perhaps

educators and educational technology researchers are

not yet convinced about this, or the cost of devices

may be a barrier. Regardless of what the reasons are,

we expect that the era of wearables in context-aware

learning environments has begun.

Location-awareness was strongly present in the

reviewed learning environments. According to our

interpretation, the reason for this is that most context-

aware learning environments are not developed to be

used in classrooms; they are informal learning envi-

ronments located beyond the physical school bound-

aries. In such systems it is essential that the learn-

ing environment can adapt its behavior to match the

learner’s whereabouts. The popularity of RFID/NFC

and GPS sensors proves this point, and it is aligned

with the popularity of the spatio-temporal context en-

tity group. Interestingly, these ﬁndings match with the

ﬁndings of our survey on context-aware learning en-

vironments published in 2009 (Laine and Joy, 2009).

In the previous survey, we also discovered that RFID

was the most common sensor technology, and pre-

dicted that RFID would become the next big thing

in wireless mobile communications. Our current re-

sults suggest that this has been the case in context-

aware learning environments, but a question remains

as to how long will the popularity of RFID/NFC as

an object identiﬁcation technology last, given the re-

cent advances in internet of things, indoor position-

ing, augmented reality and automatic object recogni-

tion through machine vision.

We can notice from Table 2 that the least used con-

text acquisition method is derivation. The other two

methods (sensors, user input) are almost equally pop-

ular with approximately 70% of the surveyed learn-

ing environments using them. We make a prediction

that in future learning environments there will be a

growing interest toward generating context data by

performing reﬁnement operations on raw sensor data.

Consquently, the popularity of derived context acqui-

sition will increase. This prediction is heralded by the

recent boom of machine learning approaches, such as

deep learning (LeCun et al., 2015), which allow so-

phisticated derivation of abstraction levels based on

raw data inputs.

As Table 2 shows, ontology and database-based

context modeling approaches were commonly used

for context modeling in the reviewed learning envi-

ronments. The main advantage of ontologies is that

they support reasoning and data content validation,

thus making them a popular and effective way to

model the context. We expect that ontologies will

keep their dominant place as a context modeling ap-

proach in context-aware learning environments in the

near future, although novel approaches based on ma-

chine learning are likely to emerge.

Given large amounts of context data acquired by

the reviewed context-aware learning environments, it

was surprising to us that there was very little con-

sideration on data security and privacy. It seems to

us that innovative use of technology and good learn-

ing outcomes have been prioritized over data safety.

A Survey of Context-awareness in Learning Environments in 2010-2016

239

Many learning environments, and mobile applications

in general, follow the learner’s location in real-time.

Will the issues of data security and privacy become

more topical when future learning environments will

be able to detect far more personal data from the

learner, such as emotions and intentions? We suggest

that these issues should be tackled sooner than later.

There are some limitations that should be consid-

ered when applying these results. Firstly, we were not

able to insert all information to Tables 1 and 2 due to

lack information in the source articles. We were also

planning to include context reasoning techniques used

in the reviewed learning environments, but only a few

studies reported about them. Secondly, although we

searched popular databases, there may be articles that

were not found during the search. Moreover, a few

articles were inaccessible due to paywalls. In spite

of these limitations, the results of this survey shed

light into contemporary context-aware learning envi-

ronments and therefore they can be useful to inter-

ested parties.

4 CONCLUSIONS

Context-aware learning is a promising research ﬁeld

that can change the way in which we learn espe-

cially in informal learning contexts. In this survey,

we compared state-of-the-art learning environments

in this ﬁeld. We provided a general overview of the

surveyed systems as well as technical details of their

context-aware architectures. In particular, we high-

lighted the most used technologies and methods for

acquiring and modeling context data. Moreover, we

proposed a taxonomy for context categories used in

context-aware learning environments. These results

could be good references for context-aware learning

environment designers and researchers who intend to

contribute to this ﬁeld.

The popular technologies and approaches identi-

ﬁed in this survey will remain to dominate context-

aware learning environments for some time, but we

can already see changes in the horizon. The fu-

ture of context-aware learning environments looks

bright given the unprecedented availability of afford-

able smart gadgets that form the Internet of Things.

These devices, together with highly sophisticated rea-

soning algorithms, will form the backbone of future

learning environments that not only serve but also

self-evolve. Our next step is to propose a conceptual

model for building future context-aware learning en-

vironments. Moreover, it is important to consider the

aspects of security and privacy, as these topics were

largely ignored by the reviewed studies.

ACKNOWLEDGMENTS

This work was supported by the Korean Na-

tional Research Foundation project NRF-

2015R1C1A1A02036469.

REFERENCES

Akbari, F. and Taghiyareh, F. (2014). E-SoRS: A person-

alized and social recommender service for E-learning

environments. In 8th National and 5th International

Conference on e-Learning and e-Teaching, ICeLeT

2014, pages 1–12.

Alharbi, M. T., Platt, A., and AI-Bayatti, A. H. (2012).

Context-aware Personal Learning Environment. In

The 7th International Conference for Internet Tech-

nology and Secured Transactions (ICITST-2012),

pages 692–697.

Baccari, S., Mendes, F., Nicolle, C., Soualah-Alila, F., and

Mahmoud, N. (2015). Comparative Study of the Mo-

bile Learning Architectures. In e-LEOT 2nd EAI Inter-

national Conference on e-Learning e-Education and

Online Training, Novedrate.

Benlamri, R. and Zhang, X. (2014). Context-aware recom-

mender for mobile learners. Human-centric Comput-

ing and Information Sciences, 4(1):12.

Bower, M. and Sturman, D. (2015). What are the educa-

tional affordances of wearable technologies? Com-

puters & Education, 88:343–353.

Capuano, N., Gaeta, M., Salerno, S., Mangione, G. R., and

Moma, S. A. (2011). An Ontology-Based Approach

for Context-Aware e-Learning. In 2011 Third Interna-

tional Conference on Intelligent Networking and Col-

laborative Systems, pages 789–794.

Chen, C. C. and Huang, T. C. (2012). Learning in a

u-Museum: Developing a context-aware ubiquitous

learning environment. Computers and Education,

59(3):873–883.

Chen, C.-C. and Lin, P.-H. (2016). Development and evalu-

ation of a context-aware ubiquitous learning environ-

ment for astronomy education. Interactive Learning

Environments, 24(3):644–661.

Chen, C.-M. and Li, Y.-L. (2010). Personalised context-

aware ubiquitous learning system for supporting ef-

fective English vocabulary learning. Interactive

Learning Environments, 18:341–364.

Chorﬁ, H. O., Sevkli, A. Z., and Bousbahi, F. (2012). Mo-

bile learning adaption through a device based reason-

ing. In Cyprus International Conference on Educa-

tional Research (CY-ICER-2012), volume 47, pages

1707–1712.

Constantino Martins, A., Faria, L., Vaz de Carvalho, C., and

Carrapatoso, E. (2008). User Modeling in Adaptive

Hypermedia Educational Systems. Educational Tech-

nology & Society, 11(1):194–207.

Gomez, J., Huete, J., and Hernandez, V. (2016). A contex-

tualized system for supporting active learning. IEEE

CSEDU 2017 - 9th International Conference on Computer Supported Education

240

Transactions on Learning Technologies, (99):196–

202.

omez, S. and Fabregat, R. (2010). Context-Aware Content

Adaptation in mLearning. In Proceedings of the 9th

World Conference on Mobile and Contextual Learn-

ing.

omez, S., Zervas, P., Sampson, D. G., and Fabregat, R.

(2014). Context-aware adaptive and personalized mo-

bile learning delivery supported by UoLmP. Journal

of King Saud University - Computer and Information

Sciences, 26(1):47–61.

Harchay, A., Cheniti-Belcadhi, L., and Braham, R. (2015).

A Context-aware Approach for Personalized Mobile

Self-Assessment. Journal of Universal Computer Sci-

ence, pages 1061–1085.

Hsu, T., Chiou, C., Tseng, J. C. R., and Hwang, G. (2016).

Development and Evaluation of an Active Learning

Support System for Context-Aware Ubiquitous Learn-

ing. Learning Technologies, IEEE Transactions on

Learning Technologies, 9(1):37–45.

Hwang, G. J. and Tsai, C. C. (2011). Research trends in

mobile and ubiquitous learning: A review of publica-

tions in selected journals from 2001 to 2010. British

Journal of Educational Technology, 42(4):65–70.

Jia, H., Wang, M., Ran, W., Yang, S. J. H., Liao, J., and

Chiu, D. K. W. (2011). Design of a performance-

oriented workplace e-learning system using ontology.

Expert Systems with Applications, 38(4):3372–3382.

Kasaki, N., Kurabayashi, S., and Kiyoki, Y. (2012). A geo-

location context-aware mobile learning system with

adaptive correlation computing methods. In The 9th

International Conference on Mobile Web Information

Systems (MobiWIS), volume 10, pages 593–600.

Kim, J.-I. and Lee, H.-H. (2014). A Intelligent English Sit-

uated Learning System based on Signboard Informa-

tion. International Journal of Multimedia and Ubiq-

uitous Engineering, 9(5):291–300.

Kitchenham, B. and Charters, S. (2007). Guidelines for

performing Systematic Literature reviews in Software

Engineering Version 2.3. Technical Report 4ve.

Laine, T. H. and Joy, M. (2009). Survey on Context-Aware

Pervasive Learning Environments. International Jour-

nal of Interactive Mobile Technologies, 3(1):70–76.

Laine, T. H. and Nygren, E. (2016). Active and passive

technology integration: a novel approach for manag-

ing technology’s inﬂuence on learning experiences in

context-aware learning spaces. Technology, Pedagogy

and Education, 25(1).

LeCun, Y., Bengio, Y., Hinton, G., Y., L., Y., B., and G., H.

(2015). Deep learning. Nature, 521(7553):436–444.

Li, M., Ogata, H., Hou, B., Uosaki, N., and Mouri, K.

(2013). Context-aware and personalization method in

ubiquitous learning log system. Educational Technol-

ogy and Society, 16(3):362–373.

Li, X., Eckert, M., Martinez, J.-F., and Rubio, G. (2015).

Context Aware Middleware Architectures: Survey and

Challenges. Sensors, 15(8):20570–20607.

Ogata, H., Yin, C., El-Bishouty, M. M., and Yano, Y.

(2010). Computer supported ubiquitous learning envi-

ronment for vocabulary learning. International Jour-

nal of Learning Technology.

Perera, C., Zaslavsky, A., Christen, P., and Georgakopoulos,

D. (2014). Context aware computing for the internet

of things: A survey. IEEE Communications Surveys

and Tutorials, 16(1):414–454.

Santos, O. C., Saneiro, M., Boticario, J. G., and Rodriguez-

Sanchez, M. (2015). Toward interactive context-aware

affective educational recommendations in computer-

assisted language learning. New Review of Hyperme-

dia and Multimedia, 22(1-2):27–57.

Scott, K. and Benlamri, R. (2010). Context-aware ser-

vices for smart learning spaces. IEEE Transactions

on Learning Technologies, 3(3):214–227.

Siadaty, M., Ga

sevi

c, D., Jovanovi

c, J., Pata, K., Miliki

N., Holocher-Ertl, T., Jeremi

c, Z., Ali, L., Giljanovi

A., and Hatala, M. (2012). Self-regulated Workplace

Learning: A Pedagogical Framework and Semantic

Web-based Environment. Educational Technology &

Society, 15(4):75–88.

Strang, T. and Linnhoff-Popien, C. (2004). A Context Mod-

eling Survey. Graphical Models, Workshop o(4):1–8.

Verbert, K., Manouselis, N., Ochoa, X., Wolpers, M.,

Drachsler, H., Bosnic, I., and Duval, E. (2012).

Context-Aware Recommender Systems for Learning:

A Survey and Future Challenges. Learning Technolo-

gies, IEEE Transactions on, 5(4):318–335.

Vladoiu, M. and Constantinescu, Z. (2011). U-Learning

within a Context-Aware Multiagent Environment. In-

ternational Journal of Computer Networks & Commu-

nications (IJCNC), 3(1):1–15.

Wang, C. S. and Wang, Y. H. (2011). Design of an SOA-

based ubiquitous learning environment. In Proceed-

ings - 2011 IEEE International Conference on Gran-

ular Computing, GrC 2011, pages 697–702.

Wang, S. L. and Wu, C. Y. (2011). Application of context-

aware and personalized recommendation to imple-

ment an adaptive ubiquitous learning system. Expert

Systems with Applications, 38(9):10831–10838.

Wu, P.-H., Hwang, G.-J., Su, L.-H., and Haung, Y.

(2012). A Context-Aware Mobile Learning System

for Supportive Cognitive Apprenticeships in Nursing

Skills Training. Educational Technology & Society,

5(1):223–236.

Yaghmaie, M. and Bahreininejad, A. (2011). A context-

aware adaptive learning system using agents. Expert

Systems With Applications, 38:3280–3286.

Yin, C., Zhang, B., Rong, W., David, B., and Xiong, Z.

(2013). Mobile learning for professional situations in

smart city. In IET Conference Publications, volume

2013, pages 401–406.

Yu, Z., Zhou, X., and Shu, L. (2010). Towards a semantic

infrastructure for context-aware e-learning. Multime-

dia Tools and Applications, 47(1):71–86.

Zhang, B., Yin, C., David, B., Xiong, Z., and Niu, W.

(2016). Facilitating professionals’ work-based learn-

ing with context-aware mobile system. Science of

Computer Programming, 1:1–17.

A Survey of Context-awareness in Learning Environments in 2010-2016

241