ArticlePDF Available

Abstract and Figures

E-Learning and online education have made great strides in the recent past. Ithas moved from a knowledge transfer model to a highly intellect, swift and interactive proposition capable of advanced decision-making abilities. Two challenges have been observed during the exploration of recent developments in e-learning. Firstly, to incorporate e-learning systems effectively in the evolving semantic web environment and secondly, to realize adaptive personalization according to the learner's changing behaviour. An ontology-driven system has proposed to implement the Felder-Silverman learning style model in addition to the learning contents, to validate its integration with the semantic web environment. Software agents are employed to monitor the learner's actual learning style and modify them accordingly. The learner's learning style and their modifications are made within the proposed e-learning system. Cloud storage is used as the primary back-end in order to maintain the ontology, databases and other required server resources. To verify the system, comparisons are made between the information presented and adaptive learning styles of the learner along with actions of agents according to learners' behaviour. Finally, various conclusions are drawn by exploring the learner’s behavior in an adaptive environment for the proposed e-learning system.
Content may be subject to copyright.
The paper has been accepted at Knowledge Based System Elsevier on 2 october 2015.
This paper can cite as: Monika Rani, Riju Nayak and O.P. Vyas An Ontology-based Adaptive
Personalized E-learning System, Assisted by Software Agents on Cloud Storage, Knowledge Based
System, Elsevier accepted on 2 october 2015.
An Ontology-based Adaptive Personalized E-learning System,
Assisted by Software Agents on Cloud Storage
AbstractE-Learning and online education have made great strides in the recent past. Ithas moved from a knowledge
transfer model to a highly intellect, swift and interactive proposition capable of advanced decision-making abilities. Two
challenges have been observed during the exploration of recent developments in e-learning. Firstly, to incorporate e-learning
systems effectively in the evolving semantic web environment and secondly, to realize adaptive personalization according to
the learner's changing behaviour. An ontology-driven system has proposed to implement the Felder-Silverman learning style
model in addition to the learning contents, to validate its integration with the semantic web environment. Software agents are
employed to monitor the learner's actual learning style and modify them accordingly. The learner's learning style and their
modifications are made within the proposed e-learning system. Cloud storage is used as the primary back-end in order to
maintain the ontology, databases and other required server resources. To verify the system, comparisons are made between
the information presented and adaptive learning styles of the learner along with actions of agents according to learners'
behaviour. Finally, various conclusions are drawn by exploring the learner’s behavior in an adaptive environment for the
proposed e-learning system.
Keywordse-learning, ontology, semantic web, cloud storage, software agents, adaptive.
1. INTRODUCTION
The emerging semantic web needs to develop an e-
learning system which focuses on personalized and
adaptive learning style of learners rather than just content
delivery. Current E-learning system of read/write web
(web 2.0) is facing some challenges to meet the
requirements of semantic web (Web 3.0). Some issues of
current e-learning system are to manage huge continuous
growing e-learning content on the internet, searching an
appropriate e-learning content as per the learner‟s
requirement, represent knowledge in machine readable
format with reasoning capability and also to allow reuse
of e-learning material. All these issues are addressed by
using ontologies for storing e-learning content and
building an e-learning application for the semantic web.
The ontology presents the course taxonomies in an
unambiguous format which is its main resolution.
Machinereadability and parsing capabilities of ontology
makes it ideal for collaborative purposes. The knowledge
base can be shared with other applications of similar
intent. In our proposed e-learning system, the detail of
personalization is stored appropriately in ontology, based
on the Felder-Silverman model [1] and dynamic changes
are notified by JADE agents. Ontology provides
personalized e-learning content as the learner‟s
requirements change dynamically and the agents capture
these changes in learning style and store this information
in the ontology. The agents collaborate to thus provide
accountability for adaptive learning. To store ontologies
we require an expanded and secure environment, thus the
entire system is deployed on DigitalOcean‟sremote cloud
host. Cloud can store incremental e-learning content and
also provides security by preventing unauthorized access
of e-learning content. By overcoming these issues we can
Monika Rani, Riju Nayak and O.P. Vyas
IIIT Allahabad
*monikarani1988@gmail.com
develop an effective and enlightening ontology driven
personalized and adaptive e-learning system.
The rest of this paper is structured into 6 sections; initially
we focus on introduction of paper in section 1. In Section
2, Foundation provides an overview of technologies used
to propose our system, like, e-learning, ontologies, cloud
computing for storing ontology and multi-agents
architecture for interaction among agents. In Section 3,
we explain the methodology details with the specification
related to the actual technologies engaged on the basis of
the foundation of the system. Methodology mentions the
technologies used like Felder-Silverman learning model,
ontology building tools and languages, DigitalOcean‟s
cloud hosting and multi-agent architecture for
development of proposed ontology driven adaptive &
personalized e-learning system. Section 4 shows
experimental results, which will affirm the content
provided to the learner by the system, by imitating to the
learner‟s learning style. It also mentions the agents‟
actions and their impact on the adaptive nature of
personalization realized by the system. Followed by
section 5, in which Learner‟s dimension, Instructor
dimension, Course dimension, Technology dimension and
the Design dimension are used to evaluate the
effectiveness of the proposed e-learning system. Finally,
section 6 draws conclusion and future research
opportunities in the current e-learning scenario.
2. FOUNDATIONS
2.1 E-learning and readiness
E-learning has gradually emerged as one of the most
frequently used technologies in the modern era. The
importance of e-learning is highlighted through
emphasizing learning techniques as well as patterns.
Hisham et al. has briefly defined it as a learning platform
that utilizes information and communication technologies
as well as electronic media. They also implied a number
of alternative terms for e-learning such as technology
enhanced learning, computer-based training, online-
education, and others [2]. This definition quite immensely
generalizes the utility of e-learning, which is of high
importance, as the scope and approach of constructing an
e-learning system is heterogeneous. Focus on a particular
design of a system may vary completely from another
design and heterogeneity in it leads to segregation of
research areas within e-learning. Various segregations
require different approaches to actualize the desired
system.
A few studies have been reviewed in order to
concisely comprehend the readiness of e-learning as a
field. A study in 2004 conducted by the U.S. Coast Guard
(USCG) focused on the validation of e-learning readiness
and was achieved internally via checking the consistency
of objects assigned to the development of self-assessment.
Data obtained from it, was later employed as a guide for
better enhancements that seemed fit for the development
of numerous instruments working towards the cause.
Respondents included members from the USCG within
the age range of 17 to 34. Despite the study focusing on
online learning, the respondents didn‟t have to actively
use any online courses. The assessment of results
confirms the potential in terms of validation and
consistency, and it also shows indication of a good
prediction tool in determining e-learning performance [3].
In 2005, Directors of Human Resource Department of,
several companies in Turkey started an initiative to assess
e-learning readiness in emerging countries. Top 100
companies in Turkey have been selected to become a part
of a survey, by the Istanbul Chamber of Industry
(ICI),from it‟s of 500 Major Industrial Enterprises of
Turkey, published in 2001. While achieving a precise
review of their e-learning readiness, they arrived at a
conclusion to develop the companies‟ HR structure before
proceeding with the integration of the online courses [4].
In 2006, a study focused to unravel the readiness among
the teachers of an institute, rather than its students, was
conducted by the academic staffs and deans of the
International Islamic University Malaysia. The study
underlined two factors, which played a crucial role in
determining their readiness, e-learning training and
confidence. However, it was suggested that their
improvement hinged upon the infrastructure of the
institute. The study also concurred that gender did not
play a factor among its respondents [5].
In 2008, a study was conducted to review the feasibility in
the health department, which was done by the nurses in
Flemish hospitals in Belgium. The analysis also
highlighted the necessity of training along with
determining the importance of strict protocols involving
work hours. It also emphasized the importance of
transparency between communication involving the
developers and people in charge of hospital policies [6].
In 2011, the focus of assessing e-learning readiness
was to candidly determine its acceptance among students
of different levels of proficiency in a subject. A group of
undergraduate students studying English as a Foreign
Language were selected from the King Khalid University
in Saudi Arabia as respondents. The study showed a
complete acceptance in e-learning integration in their
environment [7].
In 2013, a study was conducted to determine the
readiness of PhD scholars in the Christian University of
Thailand. Students were selected from various years into
their research and the aspects taken into account, while
quantifying their acceptance were technology access,
online audio/video, importance to success, internet
discussion, online skill and relationship, and motivation.
As a whole, the study uncovered a great extent of e-
learning readiness, wherein, technology access proved to
be the most promising aspect while motivation was
theleast. The difference in demographics, according to
their year in research or gender had no significance at the
readiness level [8].
In 2015, Satpute et al. reviewed several prototypes
engaged for educational needs and compared their
usability to find the advantages of using Augmented
Reality (AR). Web 2.0 tools were also examined to
understand the combined use of the two technologies.
They concluded by assertingbetter results in educational
achievements come through by combining technologies.
AR enhances immersion and engagement, whereas web
2.0 supports social interaction and collaboration [9].
2.2 Ontology Type, Specification Languages,
Development Methodologies and various application
areas:
The learner personalization details as well as the
taxonomy of learning resources will be maintained in
ontology. There have been many attempts to define an
ontology, though all of them have described the same
concept but from different perspectives. However, in
1998, Studer et al. made an attempt to define the term
while keeping in consideration, all the contemporary
perspectives and stated that “An ontology is a formal,
explicit specification of a shared conceptualization” [10],
the definition which needs a thorough explanation to
decipher a comprehensive understanding. The word
formal implies that the knowledge or content represented
by the ontology is stored in a format understood by
computers, which makes parsing the content, trivial. In
the paper, DL Query1is used as the query language to
parse the data. The explicit specification means that no
relationships or concepts depicted by the ontology can be
assumed or is implicit. Every property and relationship
must be listed in its entirety, with none left to be assumed,
which could result in multiple inferences. Finally, shared
conceptualization reflects knowledge and its constituent
conception to have a definition entirely agreed upon. This
entails that the content represented by the ontology is
universally accepted and only has a single perspective and
context to understand. In the system, domain ontology is
used, as the personalization to be demonstrated is done
sufficiently through a localized batch of concepts
pertaining to a specific domain. Elaboration is needed for
the development of domain ontology to provide
personalized e-learning.
2.2.1 Ontology definition from the prospective of e-
learning an application area:
Having a formal representation of knowledge is
helpful in interoperability within heterogeneous e-
learning environments.
Explicit specification goes towards the
enhancement of exhaustive learning by not making
assumptions on the implicit nature of the
information or the learner‟s style.
Shared conceptualization ensures that the
knowledge being stored and used, has no
ambiguity or over its definition.
Elaboration is needed for the development of domain
ontology to provide personalized e-learning. There have
been few attempts at creating domain ontologies to meet
the e-learning demands, even though its importance is
substantial.
1DL Query, Protégé wiki -
http://protegewiki.stanford.edu/wiki/DLQueryTab
There are multiple aspects to be considered while
classifying ontologies. These aspects may be
characterized according to their formality, granularity,
computational capability and generality.
2.2.2 Types of ontology in accordance to the
generality will be explained as:
Top-level ontologies [11] also referred to as upper
ontologies or foundational ontologies are domain-
independent, high level ontologies. Examples
include overly generalized, cross domain
ontologies explaining general concepts such as
Time, Space and others.
Mid-level ontologies, also known as utility
ontologies, behave as a channel between top-level
ontologies and domain ontologies. Their purpose
is similar to that of software libraries in object-
oriented programs.
Task ontologies are developed in order to store
content relevant to a specific task, like, presenting
fundamental concepts related to an overly general
activity or task.
Domain ontologies specify concepts, their
properties and relationships pertaining to a specific
domain of interest. Example ontologies showing
principal concepts which relate to a generic
domain. Therefore, the scope needs to be very
distinctly specified.
Application ontologies are created for the purpose
of aiding specific applications. They generally are
the combination of domain and task ontologies.
For instance, it includes the most specialized
ontologies which are application specific, focusing
on a definitive task or domain.
2.2.3 Ontology languages:
In this present work, various web ontology languages
are explained as the research is focused on the semantic
web. A few web ontology languages are OIL,
DAML+OIL, SHOE, XOL and OWL. Interoperability is
ensured in the web environment, as these languages are
based on the web standards XML and RDF.
Extended Markup Language (XML) is a markup
language which tries to segregate web content
from web presentation. A major drawback is its
lack of semantics, although it‟s widely used as the
web standard to represent information [12].
Resource Description Framework (RDF) is a W3C
standard used to represent web resources. A
statement in RDF is called a triple which consists
of a subject, predicate and object. A triple can be
imagined as a directed link between two nodes,
which can be modeled as the subject and the
object, whereas the predicate acts as a directed link
which is from the subject to the object. The
purpose of RDF is to allow exchange of machine-
understandable information, mainly on the web
[13].
Ontology Interface Layer (OIL)was developed
during the On-To-Knowledge project. It is
established on frame-based languages, description
logics and web standards. Its purpose is for both
representing and exchanging ontologies [14].
DAML+OILare the consequence of an effort to
merge DARPA Agent Markup Language(DAML)
and OIL. DAML+OIL show more efficiencythan
OIL due to increase in features from description
logics. However, due to the exclusion of several
frame-based features, usability with frame-based
tools became limited [15].
An XML-based Ontology Exchange Language
(XOL)is designed as a framework to exchange
ontology definitions [16].
Simple HTML Ontology Language (SHOE)extends
and allows HTML pages to incorporate machine-
readable semantic knowledge [17].
Web Ontology Language (OWL)is a standard for
representing ontologies on the semantic web.
Web-Ontology (WebOnt) Working Group
developed it in 2001. It soon became a W3C
recommendation in 2004. OWL provides
developers with a superior power to express
semantics, and to allow automated reasoners to
derive knowledge and to carry out logical
inferences [18].
2.2.4. Building ontology:Methodologies for the
development of ontologies can be traced back to the time
of development of the Cyc ontology, during which Cyc
developers published their experiences [19].
Subsequently, experience in developing the Enterprise
Ontology [20] and the TOVE (Toronto Virtual Enterprise)
[21] ontology was also reported. After these proposals, a
series of ontology development methodologies were
presented, including KACTUS [22], METHONTOLOGY
[23], Sensus [24], On-To-Knowledge [25] and CO4 [26].
In this article, the focus is on the method proposed in
the KACTUS project, in which, the ontology is built in a
bottom-up manner from a knowledge base (KB)
application. It was achieved through abstraction which
selects a KB for a specific application, and when the need
arises to create an ontology for a similar application, the
first KB abstracted represents an ontology for both
applications.
2.2.5 Ontology Applications in various fields:
Ontology based Question and Answering:
Ontology plays an important role in the
development of knowledge based system which
describes semantic relationships among entities.
These relationships described with the help of
ontology are able to reach accurate answers to a
user‟s question. Question Answering systems;
improve if the emphasis is on the semantic
analysis of literal terms in a user‟s query rather
than the syntax analysis. The Fuzzy Ontology
plays a vital role in understanding such
ambiguous user questions. Fuzzy Ontology can
help in understanding Semantic relationships by
applying Fuzzy logic (Fuzzy Type 1 and Fuzzy
Type 2) to deal with vagueness [27]. Fuzzy type-
1 can deal with Crisp membership, whereas
Fuzzy type-2 deals with Fuzzy membership.
Ontology based decision support system:
Ontology can be used along with various
methods to build intelligent discussion support
system. Ontology-supported case-based
reasoning (OS-CBR) method [28].
Ontology based E-commerce services:
Ontologies provide web services through various
heterogeneous domains by using Agent
technologies. In an environment Agent allows us
to access, retrieve and process relevant content
from various domain ontologies. Ontology
integrates with the Multi - agent platform to
improve e-commerce application for B2B and
B2C. For example, inreal world scenarios if user
wants to retrieve information like flight booking,
hotel reservation, banking transaction, etc.
information from the heterogeneous domain
ontology in single click is hard to obtain results,
as we first need to integrate these domains. The
main goal of this research is to develop a real
time system with the help of ontologies to
support systems like railways, flight ticket
booking domain, hotel booking domain, banking
transaction system, information retrieval system
etc. by communicating among agents having
domain ontology knowledge. Fuzzy ontology
with multi-agent platform system (FOMAS) to
give a proposal to automate the personalized
example for flight ticket booking domain. Fuzzy
type 2 is helpful for retrieval with multi-agent
platform system (T2FOMAS). When we
consider security aspect of such system called
secure type fuzzy ontology multi agent system
(ST2FOMAS) [29].
Ontology based Recommender system: Wu et al.
describe Fuzzy set technique can be used to
express user preference for recommendation of
items, e-books in business like e-service [30] and
e-learning.
In 2013, Sarmacakulaet al. indicate overlapping
point between e-learning and knowledge
management to provide personalized course
content. The merging of various dimensions like
student personality using Myers-Briggs learning
styles, course content is described in the form of
IEEE Learning objects (LO), TECHNOLOGY
USING FIAP device ontology and at knowledge
level using taxonomy to provide personalized
content [31-52].
2.3 Cloud Computing
In the proposed architecture cloud storage is used as
its primary back-end. The cloud is where the proposed
system is deployed along with its complementing
learner/user database, ontologies such as the user.owland
course.owl, and the relevant learning resource for the
courses. The advantages of using cloud storage can be
directly derived from many of its generic advantages.
The cost of backing up data on a cloud is
incremental, therefore saving large costs of
physical servers.
Cloud storage is said to be invisible, implying that
there isn‟t any prerequisite for a physical space to
store the servers required for the proposed system.
Security becomes unquestionable, as it is enforced
via DSA public key encryption with the remote
terminal.
Many cloud services provide several APIs to work
with by default, therefore increasing automation
from the beginning.
Cloud Storage has extensive features in its online
GUI to further ease backing up and downloading.
Accessibility of cloud storage is by default
extended to multiple platforms such as mobile
phones and tablets along with real-time syncing.
Salesforce.com marked the advent of cloud
computing. The service utilized the internet to grant
business applications in 1999. Software-as-a-Service
(SaaS) is what this is called now. In 2002, Amazon Web
Services (Amazon Mechanical Turk) was introduced,
following which, in 2006, Elastic Compute Cloud (EC2)
was launched as a commercial web service providing
computing capacity. Amazon is credited with pioneering
in pay-as-you-go services and the two services established
a firm foothold in cloud computing. With the introduction
of Web 2.0, and Google, others got into the business of
providing browser-based enterprise applications. Cloud
computing became apparent as the next step in evolution
and took over computing models such as cluster, grid,
utility, distribution and services computing [32].
Although there's limited clarity on the emerging trends
in cloud, the following two topics noticed: Technological
facets such as "elasticity", "multitenant", "resource
pooling", "computing", and "virtualization" and
Commercial faucets such as "self-service on demand",
"SaaS", and "pay-per-use" [33].
Cloud computing has received considerable attention
at both firm and industry levels from 2007. The cloud
computing environment modifies the role of the
computing stakeholders and also demands, regulatory
compliance with the infrastructure pertaining to the
location of the service provider [34]. The cloud provides
flexible IT resources and has completely changed the way
they are provided. Cloud computing on-demand resources
allow IT organizations to scale quickly as the need arises
which is required by the ever increasing business needs.
E. Aljenaa et al. propose an e-learning framework to store
rapidly developing e-learning resource on cloud as due to
its scalability, thus providing E-learning as a service
(EaaS) [35].
2.4 Multi-agent architecture
A multi-agent architecture is used in the proposed
system and a few of its properties are listed below, which
makes it advantages clear.
The primary advantage of its usage can be
narrowed down to the main agenda of the system,
which is personalization.
The multiple agents used can be categorized as
intelligent software agents which are autonomous,
specialized in their roles and are persistent entities.
Their multi-agent environment adds the
adaptability factor to personalization by
monitoring learner‟s activities to modify initial
learning style assessment from the Felder-
Silvermanmodel [36].
Agents differ from a normal program module in
their resistance towards the environment.
Software agents maintain their own code, but are
dormant in execution until triggered by the
environment.
Their power lies in the fact that they can be
modified without affecting its environment which
includes collaborating agents.
The relevance of personalization became clear through
the need for more specific materials according to learners'
preferences which in turn improved learners'
performance. Novel methodologies as well as proposed
framework were introduced to achieve personalization,
with encouraging results indicating enhanced learner
performance [37].
Deploying intelligent tutor in online education
emerged in the 90s of the last century. Sherman Alpert
and his colleagues conducted a research in 1999 on the
shift of using independent Intelligent Tutoring System
(ITS) compared to those operating on the World Wide
Web, showing both, architecture and features of the
system, support students' problem solving activities [38].
Another research conducted by Marcia Mitchell
proposed a framework for an intelligent tutoring system
that support Distance Learning (DL), which is high level
software based tutoring that has the ability to encompass a
wide variety of current DL technologies in a single
session [39].
In 2012 a new trend commenced, which focused on
deploying agents in learning, applications basedon mobile
games. Two studies were conducted autonomously, one in
English [40] and another in Japanese Kanji. The results
confirmed acceptance towards the use of the frameworks
proposed, which were analogous to the multimedia world
[41].
Dung et al. propose an architecture in 2011 which is to
represent domain ontology for e-learning course content
andagents interact on it to provide personalized e-learning
content to learners [42]. Subsequently, in 2013 Bokhari et
al. propose an architecture for interactive multi-agent
based learning system for distance education [43].
In 2015, a study applied Multi-Agent Systems (MSA)
in several research projects and software agent based
projects. With the aid of MAS methodology, a tool was
developed, known as the Agent tool, to understand the
role of MAS in the design of distributed software systems
[44].
3. METHODOLOGY
A general architecture of the system is presented in
Fig. 1. It stresses on the key technologies to be used in the
proposed system. A web-based application is built and
deployed on DigitalOcean‟s remote cloud host. All of its
required resources are stored in the cloud storage as well.
The domain ontology stored in the cloud will be referred
in the application to query and present required
information to the learner. A simple database, which is
provided by the MySQL Server running on the cloud host,
is used for authentication of the system. The actual
learning material will be available to the application via
linking their cloud URL in the domain ontology
attributes. A log file is maintained in the front end to
maintain learner activities relevant to modification in their
learning style. The proposed system is likely to function
as expected, without agents.
Fig.1. Architecture of Proposed System
However, agents are crucial to the adaptive nature of
the personalization. Domain ontologies include the
user.owl which stores the learner‟s personalization details.
The course.owl stores taxonomy of the course outline
which is eventually linked to its relevant learning
material. Agents utilize the maintained log file to provide
added personalization on top of the established learning
style derived from the Felder-Solomon questionnaire [45].
3.1 Felder-Silverman Learning Model
In 1988, Engineering Education published the Felder-
Silverman model as an article called “Learning and
Teaching styles in Engineering Education” to offer
insights about teaching and learning. The study is based
on Silverman's competence in educational psychology and
Felder's involvement during engineering days. In the next
10 years, their study winded up being the most frequently
cited paper in articles published in the Journal of
Engineering Education. In recent works, it has been used
as the most comprehensive, yet a simple learning model
referred for personalization in e-learning [36].
Fig.2. Results of the ILS questionnaire
An overview of several learning style models has been
inspected, including Myers Briggs, Gagné‟s Theory of
Learning Styles, Kolb Learning Style Learning Style
Inventory, The Ned Herrmann Whole Brain Dominance
Theory, and The Gregorc Style Delineator [46]. The
analysis led to an understanding that the Felder-Silverman
learning style model had measurable dimensions that
could directly relate to e-learning aspects. Although the
other learning models had persuasive measures and well-
founded methodologies in its determination, the Felder-
Silverman learning model has been deduced to have
learning style indexes that could be realized accurately
through the presented information [47].
2http://www.engr.ncsu.edu/learningstyles/ilsweb.html
The learning style has been demonstrated for two
reasons; firstly to capture the most important learning
style differences among engineering students and
secondly, to provide a good foundation for engineering
instructors to design a teaching approach that would
address the learning needs of all students. The second
purpose has been mapped to the personalization design of
the system. Student‟s learning styles, according to the
Felder-Silverman model, are classified into one of these
categories in each of the following four learning
dimensions: sensing or intuitive, visual or verbal, active
or reflexive and sequential or global.
The Felder-Solomon Index of Learning Styles (ILS)9
is a questionnaire developed by Richard Felder and
Barbara Solomon in 1991. The learner‟s preferred
dimension in the learning style model is determined by
the questionnaire. A total of 44 questionsareasked with a
compulsory answer.11 questions are asked for each
dimension. Each question has a possible „a‟ or „b‟ answer
that correlates to either one of the categories related to the
dimension for example the active or reflexive
dimension. The „b‟ answers are subtracted from the „a‟
answers to obtain a score which was an odd number
between -11 and 11.
Fig. 2, depicts the results obtained following the
questionnaire constructed by Felder and Solomon. It
shows all 4 dimensions categorized by their pole
characteristics over a scale ranging from -11 to 11. The
„X‟ on a number indicates the learner‟s value for that
particular dimension.
3.2 Ontology Methodology, Tools & Languages
Methodology: The methodology used is a derivation
of the KACTUS project, in which, the ontology is
developed from a knowledge base (KB) application, by
abstracting the content to a degree which is satisfactory.
In the proposed approach, the KB is textbooks or online
learning resources from which the ontology is developed
in a bottom up manner.
user.owl This ontology contains learner/user
details such as id, name, courses taken, learning
styles and such shown as Fig. 3. This ontology is
being directly linked to the authentication process.
course.owl Learning material and its module of
the course is stored in ontology as shown in Fig. 4.
Textbooks and learning materials contain the
highest degree of details. Starting from that level,
the ontology is created by abstracting thedetails,to
reach a level of classification that allowscreating a
satisfactory taxonomy.
Tools: Protégé Ontology Editor3 is used to develop
the ontology. The editor provides a GUI to achieve the
same as shown in Fig. 5. It has a highly pluggable
architecture, providing easy expansion of various utilities.
Protégé is open source software that was developed at
Stanford University in association with the University of
Manchester, and is made accessible under the Mozilla
Public License 1.1. OWL API 3.5.1 and HermiT 1.3.8
APIs used in Java to implement ontologies via
Programming and HermiT is used as a reasoner to
verifycreated ontologies. The HermiT is run to make sure
there are no inconsistencies inside the ontology.
Languages: OWL (Web Ontology Language) is used
to specify the ontology in RDF/XML format. For
querying, DL Query 2.0.2 was used which is a
representative of the Manchester Syntax4 of OWL. DL
Query is used as the query language to parse the data.
Example Fig. 6 shows the list of students attending a
lecture.
Fig.3. Learner/ user ontology (user. owl)
Fig.4. Course ontology abstracted from a KB (course.owl)
3http://protege.stanford.edu/
4http://protegewiki.stanford.edu/wiki/Manchester_OWL_Syntax
Fig.5. Protégé graphing the created ontology
Fig.6. Executed DL Query
Fig.7. Ontologies stored in DigitalOcean Host
3.3 DigitalOcean Cloud Hosting
The proposed e-learning system is deployed on the
cloud. Its support back-end resources are directly
available to it as they are hosted or stored on the cloud
as shown in Fig. 7. The agents, however, communicate
with the application through the JadeGateway API in
Java, which provides a communication channel between
the running agents and JSPs or Servlets.
DigitalOcean [48] is an easy-to-use and fast cloud
hosting service built for developers. The hoster has its
servers based mainly in New York, Amsterdam, and
San Francisco. It provides developers with a Virtual
Private Server with DNS management. It is a relatively
new player into the world of VPS cloud hosting.
However, DigitalOcean has expanded as a company
tremendously since its founding in 2011. The greatest
strides have been made in the last couple of years.
Setting it apart from many different VPS providers,
DigitalOcean provides a control panel entirely
customizable, designed with the intention of easing.
Since we are provided with a Kernel Virtual Machine
(KVM) the Linux command line can be accessed
directly with superuser privileges, simulating the use of
a computer with file system access. The most prominent
Linux distros are supplied in the aforementioned control
panel during the VPS conception. Selections include 32
bit and 64bit versions of Fedora Desktop, Ubuntu
Desktop, Arch Linux, CoreOS, CentOS, and Debian.
Utilities such LAMP, WordPress, Ruby on Rails,
Docker and Redmine are maintained as “One-click
Installs”.
DIGITALOCEAN COMMUNITY: Due to the
flexibility of the host, official documentation is
unable to cover technology compatibility and
usage with the provided KVM. Due to
this,DigitalOcean offers a Community. This is
essentially a developer-to-developer forum as
well as a recognized tutorial on several open
source topics. Due to its partnership with
Stripe, DigitalOcean is able to sponsor
Libscore to freely provide its developer
community with free access to analytics on
web development tools.
API V2: DigitalOcean has currently launched
its second version of API („API V2‟), which is
still in its BETA stage. API V2 is RESTful
implying its usage of best practices while
creating scalable web services. The API uses
OAuth authentication, which allows client
applications „secure delegated access‟ to
resources in the server by acting on behalf of a
resource owner. API V2 also supports IPv6.
LINUX DISTROS: CoreOS and FreeBSD are
two unix-based operating systems provided to
buyers to work with.
DATA BACKUP AND RECOVERY:
DigitalOcean adds reliability to the stored data
by giving thelearner two sorts of mechanisms
for data backup and recovery. Snapshots can
be taken manually of any instance of
DigitalOcean, which, however, requires that
the VPS be offline for a while. There is also an
option to turn on automatic backup as well,
which backs up an instance of DigitalOcean
periodically.
3.4 The Multi-Agent Architecture
The proposed e-learning system being designed is to
enhance personalization. Creating a system with a
learning model at its foundation provides a great deal of
personalization even without agent intervention.
However, the running agents are associated with the
system to provide adaptive personalization. Before
listing out the collaborating agents and their role in
context with the system, the JADE technology being
used to implement the agents is briefly explained.
JADE (Java Agent Development Framework)[49] is
a software framework that eases the development of
agents and applications using agents. It is compatible
with the FIPA specifications for interoperable
intelligent multi-agent systems. Therefore, its goal is to
simplify the building of FIPA compliant multi-agent
systems, while enhancing its use to contain the features
of a FIPA compliant system. To achieve the same,
JADE offers the following list of features to the agent
programmer:
Building a FIPA-compliant agent platform,
including the AMS (Agent Management
System), the DF (Directory Facilitator), and the
ACC (Agent Communication Channel). These
three agents are the default agents which are
activated automatically at the agent platform
start-up.
Fig.8. JADE environment
JADE provides a distributed agent platform as
shown in Fig. 8. A deployed agent platform can be
distributed across several hosts connected by a
network. A single Java application, and therefore
only a single Java Virtual Machine, was executed
on each host.
A single Java thread was used to run agents, and
Java events were used for lightweight
communications among agents in the same host.
FIPA-compliant naming services: start-up agents
obtain their GUID (Globally Unique Identifier)
from the platforms.
A GUI is provided which can be used to manage
several agents running on a platform or multiple
platforms as shown in Fig. 9. Generic agents can
be deployed on platforms to monitor and log files
can be usedto describe the agent's activities.
There are cases where a web interface is required with
the multiple running JADE agents. The application should
be based on JSP and Servlets. JADE offers some utility
classes that could help achieve this. The utility comes as
an API called the JadeGateway. The silver bullet for the
utility is the jade.wrapper.gateway package in the
communication. The package includes these classes:
JadeGateway
GatewayAgent
The system structure of a simple application
employing the JadeGateway can be explained via certain
keywords and a timeline of events depicting their
communication:
PingAgent exists in an agent platform.
BlackBoard is a Java bean created by the servlet
and used as a communication channel.
GatewayAgent is created by the servlet too, and it
behaves as a dispatcher. It‟s the main web
interface.
The timeline of events to explain their working:
The browser causes an event generating a POST
message.
The servlet handles it by invoking the send
message action.
A new BlackBoard object is created by the
invoked action. This Java Bean acts as the
communication channel between the
GatewayAgent and Servlet.
The GatewayAgent receives the BlackBoard
object. From the object, it extracts the recipient
and the content of the message. It then forwards
the message to the recipient.
PingAgent, which is the recipient, provides its
response to the GatewayAgent.
The response from thePingAgentis packaged and
sent back to the Servlet via theBlackBoard.
Fig.9. Remote Agent Management GUI is depicting the JADE environment
The JADE platform will be explained via the agent
instances running on it. A sequence diagram is depicted in
Fig. 10 to illustrate the agent instances as well as their
collaboration in order. However, prior to the explanation
of the sequence diagram a few words are said about the
agents used in the platform for adaptive personalization:
Agent:Monitor This agent is contacted by
JadeGateway in a behaviour waiting for a message from
the Gateway agent which tells the agent of the current
user_id in session. The behaviour is cycled in a waiting
state. Once it receives the user_id it kills the behaviour
and stores the user_id in a static variable. It contains a
tickerbehaviour as well, which periodically scans the
ontology for changed dimensions. It doesn‟t start until the
agent receives the user_id from the Gateway Agent. In
case the changed dimension found are greater than the
threshold of above +5 or below -5 the monitor agent
informs the update agent.
• Agent:Sniffer The sniffer agent employs a listener
behaviour which intercepts and redirects ACL messages
being exchanged. This is a JADE tool, agent which exists
to help developers debug their agents to see the
interchange of messages and check if they are functioning
properly. Here it will also be used to demonstrate the
sequence of message passing.
Agent:Update This agent employs a cyclic
behaviour which is in a waiting state until it receives an
INFORM ACL Message from update agent. The inform
message contains the changed dimensions and tells the
update agent about how much change is required in the
dimensions. Once it handles the change, it goes back to its
waiting state until it is prompted by the monitor agent
again.
The sequence diagram in Fig. 10 is explained below in
steps:
• The sniffer agent initiates its ListenerBehaviour to sniff
messages across the JADE platform as shown in Fig. 17.
As discussed, it is used to notice the sequence of message
transfer and lifetime of the agents.
• The website uses JadeGateway‟s API to initiate Monitor
agent‟s activity by sending it the current user_id in
session.
The monitor agent having received the user_id
periodically checks the ontology for dimension changes.
If the dimension change is above a certain limit the
monitor agent informs the update agent to change
dimensions.
• The update agent employs a cyclic behaviour in a
waiting state until it receives the INFORM ACL Message
from monitor to update the dimensions in ontology. Once
it is done, it goes back to its waiting state.
Fig.10. Sequence diagram of communicating agents
3.5 Workflow of the proposed Adaptive Personalized E-
learning:
Fig. 11.demonstrates the workflow of the proposed
Adaptive Personalized E-learning system. The Modules
of the workflow are explained as follows:
The course.owl ontology is created which is
responsible for representing the taxonomy of the
domain. The majority of the classes in that
ontology is either representative of a field or a
course. Those classes, pre-dominantly consist of
data attributes explicating the URLs of the
learning resources pertaining to particular courses.
The user.owl ontology is created to store the
learner details and its learning style indexes.
Comprehensive research was carried out on several
learning models which substantiate learning styles.
Following that, the Felder-Silverman learning style
model has been exercised, and its complementing
Felder-Solomon indexes are stored in the ontology.
The ontology also contains information about
learner‟s behaviour which can later be needed to
modify an existing style.
The OWL API of Java is used to create the ontologies. A
framework for inputting data into the ontologies is created
which allows for easy extension of the framework to
transform into a GUI based input system. The necessity of
such a framework created by the API is to extract data for
objects in the ontology and to ease application
development in later stages.
Fig.11. Workflow of Adaptive Personalized E-
learning system
An account must be registered with
DigitalOcean. A suitable KVM is chosen
according to the application requirement. In our
case, a basic system is chosen without any
extravagant specifications as our application is
purely demonstrative.
Tomcat and MySQL server are also installed on
the registered DigitalOcean host. It‟s better to
deal with the hostname rather than its IP address.
Tomcat and MySQL were also installed on the
local computer on which the app is being
developed in order to develop and deploy the
application faster. It has been provided for easier
and quicker debugging. The complete
application can be deployed on the Tomcat
container running on the cloud host.
The JADE environment which allows a FIPA
compliant multi-agent system to be created is set
up and run alongside the deployed application.
The agents are programmed on the localhost
itself and tested with a running JADE platform
on the localhost before the JADE platform was
running on the Tomcat container with the custom
agent instances needed for personalization.
The agents are programmed according to their
required application and collaboration needs.
These agents were put in containers within the
running JADE environment and use the
JadeGatewaycommunicates with the application.
4. .RESULTS AND DISCUSSION
4.1 RESULT
Once the learner fills the prerequisite Felder Solomon
questionnaire on the basis ofprovided learning style
indexes, the learner is redirected to their personalized
learning page accordingly. On the personalized pages,
there are hyperlinks which relate to behaviour that might
be indicative of changing learning style. In Fig. 12 the
„gallery view‟ icon is clicked. That triggers a Java
program to alter the ontology in a way as shown in the
figure. The „ChangeSG‟ (corresponding to the change in
Sequential-Global dimension) data property is changed
from „0‟ to -2‟ to reflect this behaviour. Once this value
goes below -5‟ or goes above „+5‟ the actual
dimensionvalue is changed this reflects adaptability in
learner behaviour.
The results observed in the expected system are shown
with the aid of screenshots of the application. Different
learning style indexes are shown beside their presented
information to highlight the dimensional differences. A
single illustration is also shown indicating the
communication between the agents responsible for
adaptive personalization.
Fig. 13 is indicative of the Active-Reflexive dimension
of a learner‟s style. The screenshot on the left of the
illustration shows an active learner. As shown in the
figure, active learners are provided with challenges
regularly, whenever applicable. However, active learners
are also provided with a Hide Challenges” option to opt
out of regular challenges. If clicked for a certain number
of times, the log file used to note a change in the
dimension value. The screenshot on the right of the
illustration shows a reflexive learner. Regular challenges
are hidden from reflexive learners. Similarly, reflexive
learners are also delivered with the All Challenges
option to view the challenges provided. If clicked up to a
certain time, the log file used to note a change in the
dimension value.
Fig.12. Learning behaviour observed in ontology
Fig.13. Represent Active-Reflexive dimension comparison for learner's learning style
Fig. 14.is indicative of the Sensing-Intuition dimension of
a learner‟s style. On the left of the illustration, sensing
learners are shown. They have regular quizzes and an
option to hide them, which goes towards changing the
learning dimensions in the log file. Similarly, on the right,
intuition learners are shown. Intuition learners have their
quizzes hidden, but given with an option of “All
Quizzes”, which allowed them to take quizzes and alter
the learning dimensions in the log file.
Fig. 15.is characteristic of the Visual-Verbal
dimension of a learner‟s style. Visual learners are on the
left of the illustration. Their primary learning style is via
videos, and if applicable, with an option of “Text
Explanation”, which when clicked enough times,
modifies the learning dimensions. Verbal learners are on
the right of the illustration. They learn via text with an
option of watching a video, if applicable. If they opt for it,
for a certain number of times, their learning dimensions
will reflect the same.
Fig.14. Represent Sensing-Intuition dimensions comparison for learner's learning style
Fig.15. Represent Visual-Verbal dimensions comparison for learner's learning style
Fig. 16.is characteristic of the Sequential-Global
dimension of a learner‟s style. Sequential learners on the
left are presented with a content view and an option to
choose a gallery view. The option has been circled in the
illustration. Selection of this option goes towards
modifying the learner dimensions. Global Learners, on
the right, are presented with a gallery view. The global
learners are provided with a content view button which
has been circled in the illustration. The selection of the
option goes towards changing the learning dimensions.
Fig.16. Represent Sequential-Global dimensions comparison for learner's learning style
Fig.17. Agent collaboration (Sniffer agent depicting communication between monitor and update)
Table.1: Describing “Active-Reflective”, ”Sensing-Intuition”, ”Visual-Verbal”, ”Sequential-Global” dimension
changes for learner‟s style (monika123):
Initial Dimension Values
Detected Change Values
Change Dimension Values
Updated Change Values
1, 3, -1, 1
0, 4, 0, -5
1, 1, -1, 1
3, -3, 0, 0
1,1,-1,1
-7,-6,3,-8
-1,-1,-1,-1
-2,-1,3,-3
-1,-1,-1,-1
11,12,16,18
3,3,5,5
1,2,1,3
3,3,5,5
0,0,-4,-1
3,3,5,5
0,0,-4,-1
3,3,5,5
-6,-4,-8,-7
1,3,3,3
-1,-4,-3,-2
Order of Values: “Active-Reflective”, ”Sensing-Intuition”, ”Visual-Verbal”, ”Sequential-Global” dimension
For updating the learner dimension Agent:Monitoring and
Agent:Updating communicate with each other.The sniffer
agent is used to demonstrate the following interaction. In
the illustration, the Agent:Monitoring informs about the
learning dimension changes as an ACL message. The
Agent:Updating acknowledges the same by a message. It
is indicated as a message in the sequence of message
passing depicted by the sniffer agent. The ACL message
informs the update agent that the dimension change has
crossed a threshold in order to change them in the
ontology. For example, Change in dimension for learner
(monika123), Active-Reflective dimension = “0”,
Sensing-Intuition dimension = “4”, Visual-Verbal
dimension = “0”, Sequential-Global dimension = “-5” of a
learner‟s style as shown in Fig. 17.
Felder-Silverman model Questionnaires provide a
quantitative method for initializing dimension values. In
our proposed e-learning model we depict a Table 1 in
which the results are seem to express the learner‟s
dimension values as shown in a before and after state,
each time the monitor agent inspects the ontology for
change. For example, learner (monika123) dimension
values change for Active-Reflective, Sensing-Intuition,
Visual-Verbal and Sequential-Global dimension are 0, 4,
0, -5 respectively as shown in Fig. 17. Each time the
change is noticed to be five or greater, the dimension is
updated by the update agent by two to maintain an odd
number. The values which are stored as data attributes in
the ontology are updated since the changes have been
incorporated in the actual dimension values, in order to
provide adaptive and personalized learning for learners.
4.2 DISCUSSION
The discussion articulates the meaning of the results
presented. As shown in Fig. 13. Active learners are
regularly provided with challenges, since those learners
liked to try out what they‟ve learnt. Challenges provide a
platform to exercise the knowledge. Reflexive learners on
the other hand, are not regularly provided with
challenges; as such learners would rather have time to
think over the knowledge learnt, instead of practicing
exercises based on them, immediately. Providing them
with alternate options helps the system adapt to their
changing behaviour and allow adaptability.
Fig. 14.shows a comparison between sensing and
intuition learners. Sensing learners are very particular
about the facts and details of a particular topic. It allows
the learners to fully understand the topic. Intuition
learners on the other hand, are more interested in theories
and principles, and less concerned about the details
involving the topic. Therefore, quizzes are provided for
sensing learners and are hidden from intuition learners.
The option of hiding and revealing the quizzes to sensing
and intuition learners, respectively, helps provide
alternatives and monitor their behaviour.
Visual and Verbal learners are shown in Fig. 15.
Visual learners absorb better with pictures and an audio
visual environment. Due to this reason they are primarily
provided with a video comprising of a higher number of
pictorial aids and audio explanations. They have an option
to look at the text explanations as well, signalling the
system a potential change in behaviour. Verbal learners
are given text explanations by default as they are better
learners when provided with verbal explanation rather
than diagrams. Verbal learners also have an option to
watch the video if they aren‟t completely satisfied with
the explanation or for a more complete learning. If opting
for a certain number of times the system alters the
dimensions accordingly.
The Sequential and Global learners are shown via Fig.
16. Sequential learners are comfortable with a linear
learning style in which knowledge is presented
incrementally. Due to this, a content layout is provided to
learners, which shows the order of learning clearly.
However, if they want they can switch to a gallery view to
understand the entire content as a whole. Global learners
are provided the gallery view by default, because they
preferred to look at the entire content as a whole to
understand a given topic. If they opt to look at the order
of the content they are allowed to switch to a content
view. These alternatives are noticed by the log file to
adapt to their changing behaviour.
As shown in Fig.17, Agent:Update sends the
dimension changes to the Agent:Monitor periodically to
maintain a real time adaptive system. Whenever sufficient
changes are present in the log filesAgent:Monitoring does
the required activities to make the changes permanent in
the ontology.
Such a proposed system has several applications in the
real world. Personalization not only includes objectifying
the learner‟s styles, but to also monitor learner‟s usage of
the system to conform to the current learning styles. Such
systems can help learners maintain focus with their
changing patterns. This focus can improve productivity
not only in educational institutions, but also in industrial
training and learning. Integrating with the semantic web
helps with the reusability of the system, providing a valid
milestone for further research in the current scheme of
study. Keeping the Felder-Silverman model as the
foundation of the study, it helps to keep up-to-date with
the latest learning styles. It can be credited to the fact that
the learning model is heavily researched in statistical
domains and practical situations.
Table.2: The evaluation questionnaire survey for
proposed E-learning system
Dimension
Question to ask learners
Learner
dimension
Q1. How familiar are you with an e -
learning platform?
Q2. The attitude towards the use of
computer/laptop/mobile for e-learning
purpose?
Q3. Your understanding about the
provided e-learning course content?
Instructor
dimension
Q4. How instructor organizes the
content to meet learner‟s objectives?
Q5. How timely e-learning content is
updated on the e-learning system, by
the instructor?
Q6. The amount of time given for
preparation of the quiz?
Course
dimension
Q7. How engaging and personalized
was the course content?
Q8. How relevant were the topics
covered in the course?
Q9. Was the Course content direct and
comprehensible?
Design
dimension
Q10. The interface design of an e -
learning system?
Q11. Quality of video, audio and text
used as e-learning material?
Q12. Responsiveness of the e-learning
system?
Technology
dimension
Q13. Learning management system
setup?
Q14. How well did the Felder-
Silverman Learning Model determine
your learning style?
Q15. Rate the e-learning platform on
the basis of the following
characteristics:
i. Active-Reflexive dimension
ii. Sensing-Intuition dimension
iii. Visual-Verbal dimension
iv. Sequential-Global dimension
5. EVALUATION OF PROPOSED E-LEARNING
SYSTEM
5.1 Evaluation Criteria
In 2010, Lockee et al. proposed an approach for
determining qualityof e-learning system called
“openECBCheck”[50]. The validation of our proposed e-
learning system has been done through questionnaire
survey based on dimensional factors. Sun et al. [51]
proposed six dimensions influencing learner‟s satisfaction
viz., Environmental dimension: Feedback from learner in
the form of questionnaire survey Learner dimension: Q1
to Q3, Instructor dimension: Q4 to Q6, Course dimension:
Q7 to Q9, Design dimension: Q10 to Q12 and
Technology dimension: Q13 to Q15 as shown in Table 2.
5.2 Limitation for Questionnaires surveys for proposed e-
learning system:
Limitations for leaner and research scholar from both
perspectives:Here, learner is a person who responds to the
questionnaire and research scholar is a person who
prepares the survey questionnaire
Sample size is limited due to targeted population
as it depends on the content provided for e-
learning.
Survey form can miss some questions that need
to be answered by the learner/user.
Learner can interpret questions in different
scenarios
Learner can interpret the question from different
perspectives.
The e-learning content store in course.owl will
not be applicable to other domains because the
proposed e-learning system will cover only
computer science e-learning content for now,
which can‟t be understood by another domain
leaner/user.
Fig.18. Average score
5.3 The Effective Evaluation of Proposed E-learning
system
In our proposed e-learning system, we are considering
feedback of questionnaire survey as Environmental
dimension. The questionnaire survey consists of 15
questions with a score value of a question ranging from 1
to 5. The evaluation dimensions of our system are:
Learner dimension,
Instructor dimension, Course dimension, Technology
dimension and Design dimension on the basis of which
Average score is calculated for each dimension. These
evaluation dimensions are rated on a scale of 1 to 5 points
and the results are displayed in the bar chart as shown in
Fig. 18. For “Learner dimension” the average score is
3.87. Similarly, “Instructor dimension” is rated as 3.45,
”Course dimension” is rated as 3.74, “Technology
dimension” is rated as 3.45 and “Design dimension” is
rated as 3.659.
6. CONCLUSION & FUTURE WORK
The proposed system integrates an e-learning
application with the semantic web via domain ontology.
The ontologies created for the system not only provide
re-usable content for the future applications with similar
purposes, but also represents the concept hierarchy
structure as a domain ontology with more expressive
relations. The standards provided by Ontology web
language (OWL) have made it trivial to understand the
semantics of the knowledge base. In the proposed
system, we have maintained two ontologies namely, the
course.owl and user.owl, which are used to store
learning materials and to implement the Felder-
Silverman learning style model respectively. The
Felder-Solomon Index of Learning Styles (ILS) is a list
of questionnaires which are used to exercise the
learner‟s learning preferences and present the
information to the learner accordingly. The learning
style model lays the foundation of a personalized
environment, focusing on the learner‟s pattern.DL
Query also provides an efficient way of extracting
required information from the application's ontology.
HermiT reasoner is used to determine the consistencies
in user.owl and course.owl ontologies
The collaborative software agents notice to alter the
learner‟s dimension values according to the learner‟s
usage of the application. Contradicting learning
behaviour is noticed by the agents and the preference is
changed accordingly. Adaptability has been realized by
deploying JADE agents, namely Agent:Updating and
Agent:Monitoring. The Communication between the log
file of the application and the user.owl via the
JadeGateway, is the learner‟s changing patterns which
are monitored and modified in real-time. Ontologies
deployed on DigitalOcean‟s remote cloud host provide
an expanded and secure environment to proposed e-
learning system. Therefore, the final a survey was
conducted to explore an adaptive, personalized e-
learning application using ontology to integrate with the
semantic web cloud services to, employ an incremental
model and a multi-agent system to recognize
adaptability in the learner‟s behaviour.
Future studies can be conducted to objectify the
learner‟s personalization in a better way by considering
Learner Instructor Course Design Technology
0
0.5
1
1.5
2
2.5
3
3.5
4
Evaluation Dimension
Score
the exact value of a particular dimension in the Felder-
Solomon learning style index. Lecturers can be
actualized in the system by ascertaining their teaching
style by observing learners. Personalized e-learning
course and its contents using adaptive Learning Path
Sequence (LPS) can be recommended for learners.
Evaluation parameter “Environmental dimension”can be
improved by deploying proposed e-learning system on
discussion forums, polling e-mail based tool, chat and
Instant Messaging (IM) and audio & video
conferencing. Assessing readiness of our proposed e-
learning in IOE (Internet of Everything) scenario. Our
proposed approach provides a motivation for
advancedlearning by simultaneously supporting the
vision of IOE for higher education among learners.
Enhancement in technology leads to the opportunitiesfor
the disabled learners by providing training in a better
way. Even disabled learners can be targeted in IOE
scenario for advanced digital education with various
features like cloud‟s store's ontology, where agents
interact to provide adaptive and personalized learning.
REFERENCES
[1] S. Graf, S. R. Viola, T. Leo, and Kinshuk. "In-depth
analysis of the Felder-Silverman learning style
dimensions." Journal of Research on Technology in
Education 40, no. 1 (2007): 79-93.
[2] H. Hashim and Z. Tasir, E-Learning Readiness: A
Literature Review, In Teaching and Learning in
Computing and Engineering (LaTiCE), International
Conference, IEEE, (2014): 267-271.
[3] R.Watkins, D. Leigh, and D. Triner, Assessing
readiness for e-learning, Performance Improvement
Quarterly 17, no. 4 (2004): 66-79.
[4] C. H. Aydin, and D. Tasci, Measuring readiness for e-
learning: reflections from an emerging
country, Educational Technology and Society 8, no. 4
(2005): 244-257.
[5] K. A. Abdulhameed, Assessing the awarenes and
perceptions of academic staffs in using e-learning
tools for instruction delivery in a post-secondary
institution: A case study”, The Innovation Journal:
The Public Sector Innovation Journal 11, no.4 (2006).
[6] J. Schreurs, G. Sammour, and U. Ehlers, ERA-E-
Learning Readiness Analysis: A eHealth Case Study
of E-Learning Readiness, In The Open Knowlege
Society. A Computer Science and Information
Systems Manifesto, Springer Berlin Heidelberg,
(2008): 267-275.
[7] I. F. Abdulaziz, EFL students‟ readiness for e-
learning: factors influencing e-learners acceptance of
the Blackboard in a Saudi University, Jalt Call Journal
7, no 1, (2011): 19-42.
[8] J. Wattakiecharoen, and P. Nilsook, E-learning
readiness of PhD students, International Conference
on Excellent Innovation for Educational Reaseach and
IT Learning in the 21st Century, (2013).
[9] S.Trupti, S. Pingale, and V. Chavan, Augmented
reality in E-Learning Review Of Prototype Designs
For Usability Evaluation, in Communication,
Information & Computing Technology (ICCICT),
International Conference, (2015): 1-4.
[10] S. Staab, R. Studer, H.P. Schnurr, and Y. Sure,
Knowledge processes and ontologies, IEEE Intelligent
systems 16, no. 1 (2001): 26-34.
[11] M. Hadzic, P. Wongthongtham, T. Dillon, and E.
Chang, Ontology-Based Multi-Agent Systems Studies
in Computational Intelligence, Springer-Verlag Berlin
Heidelberg. (2009).
[12] T. Bray, J. Paoli, C.M. Sperberg-McQueen, and E.
Maler, Extensible Markup Language (XML) 1.0,
Second Edition, W3C Recommendation,
http://www.w3.org/TR/REC-xml, (2000).
[13] O. Lassila, and R.Swick, Resource Description
Framework (RDF) model and syntax specification,
W3C Recommendation, http://www.w3.org/TR/REC-
rdf-syntax/, (1999).
[14] I. Horrocks, D. Fensel, F. Harmelen, S. Decker, M.
Erdmann, and M. Klein, OIL in a Nutshell, in
ECAI‟00 Workshop on Application of Ontologies and
PSMs, Berlin, (2000).
[15] I. Horrocks, and F. van Harmelen, Reference
Description of the DAML OIL Ontology Markup
Language, Technical report,
http://www.ai.sri.com/pkarp/xol/xol.html, (2002).
[16] R. Karp, V. Chaudhri, and J. Thomere, XOL: An
XML-Based Ontology exchange Language, Technical
Report, http://www.ai.sri.com/pkarp/xol/xol.html,
(1999).
[17] S. Luke, and J. Hein, “SHOE 1.01, Proposed
Specification, SHOE Project Technical report”,
University of Maryland,
http://www.cs.umd.edu/projects/plus/SHOE/spec1.01.
htm, (2000).
[18] M. Dean, D. Connolly, F. van Harmelen, J. Hendler, I.
Horrocks, D. McGuinness, P.F. Patel-Schneider, and
L.A. Stein, OWL Web Ontology Language 1.0
Reference, W3C Working Draft,
http://www.w3.org/TR/owl-ref/, (2002).
[19] D.B. Lenat, and R.V. Guha, Building Large
Knowledge-Based Systems:Representation and
Inference in the Cyc Project, Addison-Wesley,
Boston, (1990).
[20] M. Uschold, and M. King, Towards a Methodology
for Building Ontologies, In Proceedings of the
IJCAI95 Workshop on basic Ontological Issues in
Knowledge Sharing, Montreal, (1995).
[21] M. Gruninger, and M.S. Fox, Methodology for the
design and evaluation of ontologies, in Proceedings of
the Workshop on Basic Ontological Issues in
Knowledge Sharing Montreal, (1995).
[22] G. Schreiber, B. Wielinga, and W. Jansweijer. The
KACTUS view on the „O‟word, In IJCAI workshop
on basic ontological issues in knowledge sharing,
(1995): 159-168.
[23] M. Fernandez Lopez, A. Gomez Perez, and N. Juristo,
METHONTOLOGY: From Ontological Art towards
Ontological Engineering, in Proceedings of the AAAI
Symposium on Ontological Engineering, Stanford,
(1997).
[24] B. Swartout, R. Patil, K. Knight, and T. Russ. Toward
distributed use of large-scale ontologies, In Proc. of
the Tenth Workshop on Knowledge Acquisition for
Knowledge-Based Systems, (1996):138-148.
[25] S. A. Rădulescu, A Perspective on E-learning and
Cloud Computing, Procedia-Social and Behavioral
Sciences 141 (2014): 1084-1088.
[26] J. Euzenat, Corporate memory through cooperative
creation of knowledge bases and hyper-documents,
In Proc. 10th workshop on knowledge acquisition
(KAW), Banff (CA), (1996): 1-18.
[27] M. Rani, M. K. Muyeba, and O. P. Vyas. "A hybrid
approach using ontology similarity and fuzzy logic for
semantic question answering." In sAdvanced
Computing, Networking and Informatics-Volume 1,
pp. 601-609. Springer International Publishing,
(2014):601-609
[28] K. Amailef, and J. Lu. "Ontology-supported case-
based reasoning approach for intelligent m-
Government emergency response services."Decision
Support Systems 55, no. 1 (2013): 79-97.
[29] A. C.Bukhari, and Yong-Gi Kim. "Integration of a
secure type-2 fuzzy ontology with a multi-agent
platform: A proposal to automate the personalized
flight ticket booking domain." Information
Sciences 198 (2012): 24-47.
[30] D. Wu, G. Zhang, and J. Lu. "A fuzzy preference tree-
based recommender system for personalized business
to-business e-services."Fuzzy Systems, IEEE
Transactions on 23, no. 1 (2015): 29-43.
[31] Bijalwan, Vishwanath, et al. "Machine learning
approach for text and document mining." arXiv
preprint arXiv:1406.1580 (2014).
[32] D. Reeves, D. Blum, R. Watson, G. Cress, B. Blakely,
C. Haddad, and J. Lewis, Cloud Computing:
Transforming IT,
https://www.gartner.com/doc/1405582/cloud-
computingtransforming-it, (2009).
[33] R. Buyya, C.S. Yeo, S. Venugopal, J. Broberg, and I.
Brandic, Cloud computing and emerging IT platforms:
Vision, hype, and reality for delivering computing as
the 5th utility, Future General Computing Systems 25,
no 6 (2009): 519-616.
[34] N. Sultan, and S. van de Bunt-Kokhuis, Organisational
culture and cloud computing: coping with a disruptive
innovation, Technology Analysis & Strategic
Management 24, no. 2 (2012): 167-179.
[35] E. Aljenaa, F. S. Al-Anzi, and M. Alshayeji.
"Towards an efficient e-learning system based on
cloud computing." In Proceedings of the Second
Kuwait Conference on e-Services and e-Systems,
ACM, (2011): 13.
[36] R. M. Felder, and L. K. Silverman, Learning and
teaching styles in engineering education, Engineering
education 78, no. 7 (1988): 674-681.
[37] C.M. Chen, and S.H. Hsu, Personalized Intelligent
Mobile Learning System for Supporting Effective
English Learning, Educational Technology &
Society 11, no. 3 (2008): 153.
[38] S.R. Alpert, M.K. Singlev, and P.G. Fairweather,
Deploying intelligent tutors on the web: An
architecture and an example, International Journal of
Artificial Intelligence in Education 10, no. 2 (1999):
183-197.
[39] M.T. Mitchell,An architecture of an intelligent
tutoring system to support distance learning,
Computing and Informatics 26, no. 6 (2012): 565-576.
[40] J.S. Hey, H.F. Lu, J.C. Hwang, S.M Lin, and K.C. Li,
A Model for Role-Playing Game with Mobile
Learning, In Digital Game and Intelligent Toy
Enhanced Learning (DIGITEL), 2012 IEEE Fourth
International Conference IEEE, (2012): 102-104.
[41] M.B. Syson, R.E. Estuar, and K.T. See, ABKD:
Multimodal Mobile Language Game for Collaborative
Learning of Chinese Hanzi and Japanese Kanji
Characters In Proceedings of the 2012
IEEE/WIC/ACM International Joint Conferences on
Web Intelligence and Intelligent Agent Technology 3,
(2012): 311-315.
[42] P. Q. Dung, and Adina Magda Florea. "An
architecture and a domain ontology for personalized
multi-agent e-learning systems." In Knowledge and
Systems Engineering (KSE), 2011 Third International
Conference on, IEEE, (2011): pp. 181-185.
[43] M. Bokhari, and Sadaf Ahmad. "Design for Interactive
E-learning Based upon Multi-Agent System: i-
MBLS." (2013): 9-05.
[44] K. P. Anand, and R. Pandey, Role of Multi Agent
System Methodology in System Design, in 2nd
International Conference in Computing for
Sustainable Global Development (INDIACom),
(2015): 438-441.
[45] R.M. Felder, and J. Spurlin, Applications, reliability
and validity of the index of learning styles,
International journal of engineering education 21, no.
1 (2005): 103-112.
[46] L.V. Harris, M.A. Sadowski, and J.A. Birchman, A
comparison of learning style models and assessment
instruments for university graphics educators,
Engineering Design Graphics Journal 70, no. 1 (2009).
[47] R. V. Silvia, S. Graf, and R. Leo, Analysis of Felder-
Silverman Index of Learning Styles by a Data-Driven
Statistical Approach, in eighth IEEE International
Symposium on Multimedia, (2006) : 959-964.
[48] S. Shi, L. Xu, H. Dong, L. Wang, S. Wu, B. Qiao, and
G. Wang. "Research on data pre-deployment in
information service flow of digital ocean cloud
computing." ActaOceanologicaSinica 33, no. 9
(2014): 82-92.
[49] T. M. K. Nguyen, Developing an Intelligent Multi-
Agent System based on JADE to solve problems
automatically, International Conference Systems and
Informatics (ICSAI), (2012): 684-690.
[50] B. Lockee, R. Perkins, K. Potter, J. Burton, and S.
Kreb. "Defining quality in distance education:
Examining national and international standards in
online learning." In 27th Annual Conference on
Distance Teaching and Learning, Madison, USA,
(2011).
[51] Sun, Pei-Chen, Ray J. Tsai, Glenn Finger, Yueh-Yang
Chen, and DowmingYeh. "What drives a successful e-
Learning?An empirical investigation of the critical
factors influencing learner satisfaction." Computers &
education 50, no. 4 (2008): 1183-1202.
[52] S. Cakula and MaijaSedleniece. "Development of a
personalized E-learning model using methods of
Ontology." Procedia Computer Science 26 (2013):
113-120.
... An ontology-driven approach incorporating the Felder-Silverman learning style model shows promise in resolving these difficulties by allowing for the customisation of learning content within the semantic web framework [43]. This unique method ushers in a new era of Tanzanian higher education, generating more immersive and compelling learning environments that meet the different requirements of learners [44,45]. ...
Article
Full-text available
This study examines the challenges and students' perceptions of adopting and implementing e-learning in Tanzanian universities to facilitate the shift towards personalised learning. Although technology and online courses provide chances to customise training for unique student requirements, challenges impede the successful incorporation of e-learning in Tanzanian higher education, particularly regarding user training and technical support. The absence of crucial amenities for online education, such as computer laboratories and dependable internet connectivity, exacerbates the challenges in executing implementation endeavours. We employed a mixed-methods approach to investigate universities' preparedness for implementing e-learning and evaluate the impact of e-learning platforms and Artificial Intelligence (AI) in facilitating personalised learning experiences. Our findings emphasise the significance of tackling identified obstacles to enhance education quality and provide a basis for customised learning. Subsequent studies should focus on filling the existing voids to improve the incorporation of artificial intelligence in e-learning platforms, explicitly addressing the obstacles faced by universities in underdeveloped nations. The study suggests that institutions should allocate resources towards developing e-learning infrastructure, offer extensive training for instructors, and create customised e-learning methods.
... This example and other studies (Afzal et al. 2019;Gan et al. 2019;Rani, Nayak, and Vyas 2015) investigating AI-powered tutors are founded on the notion of adapting content based on personal characteristics. Whilst undoubtedly useful in providing content based on a learner's aptitude in the context of e-learning platforms, they lack the direct visual element that innovative technologies like human activity recognition (HAR) and 'digital twinning' can provide, especially in the learning of hands-on skills within fields like sport, manufacturing, art, mechanical engineering, maintenance, etc. ...
Article
With the growing integration of artificial intelligence (AI) in education, understanding its pedagogical implications has become an imperative. Unpacking Pedagogical Potential (UPP) is a study that addresses this need by exploring the efficacy of artificially generated lesson plans in the context of a modern vocational programme. The research aims to evaluate the quality and comprehensibility of these AI-driven lesson plans and teachers’ experiences. Associated benefits and challenges will also be identified. Data will be collected through a grounded theory approach, including interviews, focus group discussions, and performance evaluations. The objective is to present a holistic view of the pedagogical potential of AI in lesson planning, ultimately guiding future enhancements for AI applications in educational contexts. In the first phase of the study, two lesson plans were generated on the challenging topics of interfaces and abstraction in object-oriented programming. The lesson plans were generated using OpenAI’s GPT-4 through a one-shot prompt detailing the template to be used. Following generation, the lessons plans were peer-reviewed and evaluated using Lesson Plan Analysis Protocol (LPAP). Analysis of the score and a critical narrative of generation process was performed. Results show that even though lesson plans were generated by ChatGPT, the lesson plans require extensive human intervention to make the activities viable for use in the classroom. On the other hand, using them to structure the information presented is a viable way to ensure that the teacher follows a pre-established structure in generating a lesson plan. This can serve to therefore improve the quality of lesson plans generated by experienced teachers based on a standard metric.
... Ontology based decision methods leverage greater understanding of semantic relationships by applying probability logic e.g. fuzzy type 1 and fuzzy type 2to tackle vagueness [4], [29]. This emphasises the important role of ontologies in the development of knowledge based system which describes semantic relationships among entities [30]. ...
Preprint
Full-text available
Objects or structures that are regular take uniform dimensions. Based on the concepts of regular models, our previous research work has developed a system of a regular ontology that models learning structures in a multiagent system for uniform pre-assessments in a learning environment. This regular ontology has led to the modelling of a classified rules learning algorithm that predicts the actual number of rules needed for inductive learning processes and decision making in a multiagent system. But not all processes or models are regular. Thus this paper presents a system of polynomial equation that can estimate and predict the required number of rules of a non-regular ontology model given some defined parameters.
... This proves to be crucial as the smart-cities project must be replicated in a large number of Indian cities for it to be beneficial for everyone. Gainful information can be extracted from the ontology [12] with ease, after different departments and their scope, relations and information are clear and evidently laid out by the ontology, with the use of a querying language. ...
Preprint
Several challenges are faced by citizens of urban centers while dealing with day-to-day events, and the absence of a centralised reporting mechanism makes event-reporting and redressal a daunting task. With the push on information technology to adapt to the needs of smart-cities and integrate urban civic services, the use of Open311 architecture presents an interesting solution. In this paper, we present a novel approach that uses an existing Open311 ontology to classify and report non-emergency city-events, as well as to guide the citizen to the points of redressal. The use of linked open data and the semantic model serves to provide contextual meaning and make vast amounts of content hyper-connected and easily-searchable. Such a one-size-fits-all model also ensures reusability and effective visualisation and analysis of data across several cities. By integrating urban services across various civic bodies, the proposed approach provides a single endpoint to the citizen, which is imperative for smooth functioning of smart cities.
... The system aimed to engage students in a personalized, creative, and flexible approach to selfeducation, promoting their activity and collaboration. Moreover , the system was designed to be open for extensions and experiments with prototypes, such as service and agent-oriented architectures, aligning it with the interactive, reactive, and proactive educational processes seen in other systems [4], [5], [6]. ...
Article
Full-text available
The objective of the research is to enhance the functionality of the FraudDetector software agent within the Distributed eLearning Center (DeLC), a platform providing extensive support for e-learning activities. DeLC assists students and teachers in organizing learning materials, addressing knowledge gaps, conducting exams, and fostering personalized elearning environments. The project scope encompasses various extensions, including an agent-oriented environment that enriches functionalities with reactive and proactive intelligent components, referred to as agents or assistants. This paper focuses on the latest evolution of the FraudDetector software agent, transitioning from its base functionality for fraud detection to leveraging artificial intelligence (AI) capabilities. The goal is to integrate AI, specifically the knowledgebase provided by ChatGPT, to enhance FraudDetector's effectiveness. This integration is the primary contribution of the research, aimed at improving fraud detection precision. Experimentation reveals promising results, suggesting that involving ChatGPT enriches FraudDetector's functionality and enhances the agent's precision. Moving forward, the agent's architecture should remain open for collaboration with external AI providers, with efforts to decouple components responsible for integration. The real-world implementation of these findings is pending, warranting further validation through production environment testing.
... Early research on online teaching concentrated on the technologies and system implementations involved. Rani M primarily employed the Felder-Silverman learning style model, the LT-SA model, and the SECTIONS framework to develop various online learning systems [17]. Some scholars enhanced system efficiency by optimizing service level agreement (SLA) parameters in online teaching platforms through algorithms [18]. ...
Article
Full-text available
In the new normal, online education has become a norm and is considered equally important as traditional face-to-face education. Although online teaching has improved accessibility, reduced management costs, and expanded educational opportunities, there are also some issues, such as whether technology and internet access are widespread and whether the quality of online education is better or worse compared to previous modes. With the rapid development of online education, it is crucial to investigate the key factors influencing university teachers’ continuous usage of online teaching platforms and their working mechanisms, which is of great significance for improving the level of educational informatization and promoting educational modernization. This study mainly explores the continuous usage intention of online teaching platforms by Chinese university teachers under the new normal and its influencing factors, aiming to provide theoretical support and practical insights for improving educational informatization, promoting educational modernization, and designing and optimizing online teaching platforms. This study builds a new theoretical framework based on the technology acceptance model (TAM), UTAUT model, and the theory of planned behavior (TPB). From April to November 2024, 343 questionnaires were collected from Chinese university teachers. Combining structural equation modeling (SEM) and fuzzy-set qualitative comparative analysis (fsQCA), the key factors influencing teachers’ continuous usage of online platforms and their configuration relationships are systematically discussed. Not only are the key factors verified, but the configuration relationships are also revealed, providing deeper insights for universities to design, promote, and optimize online teaching platforms and offering theoretical support and practical guidance for promoting sustainable educational development.
... In these circumstances, AI has a significant impact on the development and implementation of clean energy sources. AI is primarily focused on developing intelligent machines and software for specific problems in various fields, one of which is the energy sector [12]. AI also plays a substantial role in advancing technologies in the renewable energy domain. ...
Article
Full-text available
This study investigates the relationship between artificial intelligence (AI), industrial robots, and renewable energy consumption, driven by the rapid technological advancements and widespread adoption of AI tools in various industries. This research aims to evaluate the environmental implications of these technologies, specifically their impact on renewable energy usage. Employing a comprehensive analytical framework, this study utilizes advanced methodologies, including regularization factors, to accurately estimate the effects of these variables. Through a thorough data analysis, the research quantifies how AI and industrial robots influence the shift towards renewable energy sources. The findings reveal that investments in AI significantly enhance renewable energy consumption, as demonstrated by both conventional estimation techniques and those that integrate regularization factors. Conversely, the use of industrial robots is found to have a detrimental effect on renewable energy consumption. These results have important implications for policymakers, industry leaders, and sustainability researchers. This study encourages policymakers and investors to prioritize funding for AI solutions that promote renewable energy adoption, while it advises industry managers to strategically modify their use of industrial robots to reduce their environmental impact. Ultimately, this research lays a critical foundation for future inquiries and policy initiatives aimed at aligning technological advancements with sustainable energy practices.
Chapter
The integration of artificial intelligence (AI) algorithms and models in the realm of embodied agents, such as drones and robots, has emerged as a transformative field with profound implications for various industries. This chapter provides a comprehensive overview of the state-of-the-art AI techniques employed to augment the autonomy, adaptability, and efficiency of these autonomous systems. The study begins by examining the fundamental principles that underpin AI algorithms for embodied agents, emphasizing the importance of sensor fusion, perception, and decision-making processes. It explores the symbiotic relationship between machine learning and embodied agents, showcasing how models evolve over time to enhance adaptability in dynamic environments. One of the key focal points is the application of deep learning architectures, including convolutional neural networks (CNNs) and recurrent neural networks (RNNs), to bolster the perceptual capabilities of drones and robots. These architectures enable real-time object recognition, scene understanding, and spatial mapping, facilitating improved navigation and interaction within complex surroundings. The chapter delves into the significance of reinforcement learning (RL) in training embodied agents, elucidating how these algorithms enable agents to learn from experiences and optimize their decision-making processes. Specifically, it investigates the challenges of implementing RL in the context of drones and robots, addressing issues related to sample efficiency, exploration–exploitation trade-offs, and safety concerns. Furthermore, the integration of AI algorithms for collaborative multi-agent systems is explored, emphasizing the coordination and communication strategies essential for achieving collective goals. The research investigates swarm intelligence, where groups of embodied agents work together seamlessly, inspired by natural systems like ant colonies and flocking behaviour in birds. The ethical considerations and societal implications of deploying AI-powered embodied agents are also discussed. As these systems become increasingly autonomous, concerns surrounding accountability, transparency, and the potential impact on employment and privacy are scrutinized. Finally, the chapter highlights emerging trends and future directions in AI for embodied agents, including the fusion of symbolic reasoning and deep learning, the integration of edge computing for real-time processing, and the continual evolution of AI models through online learning paradigms. This chapter provides a comprehensive synthesis of the current landscape of AI algorithms and models for embodied agents, shedding light on the advancements, challenges, and potential future developments in this rapidly evolving field.
Preprint
Ontologies provide features like a common vocabulary, reusability, machine-readable content, and also allows for semantic search, facilitate agent interaction and ordering & structuring of knowledge for the Semantic Web (Web 3.0) application. However, the challenge in ontology engineering is automatic learning, i.e., the there is still a lack of fully automatic approach from a text corpus or dataset of various topics to form ontology using machine learning techniques. In this paper, two topic modeling algorithms are explored, namely LSI & SVD and Mr.LDA for learning topic ontology. The objective is to determine the statistical relationship between document and terms to build a topic ontology and ontology graph with minimum human intervention. Experimental analysis on building a topic ontology and semantic retrieving corresponding topic ontology for the user's query demonstrating the effectiveness of the proposed approach.
Article
Full-text available
The purpose of this study was to recognise the perceptions of undergraduate students in different levels of the English department as to their readiness to accept e-learning as the dominant learning paradigm at King Khalid University (KKU), by probing the opinions of a sample of the EFL students. Descriptive research, involving a survey and in-depth interviews, was utilised to recognise and analyse teachers and students’ perceptions of their attitudes towards e-learning, their readiness to accept it, and the factors facilitating or inhibiting e-learning. Recognition of these factors for building and applying the instruments in this study was done based on an extensive review of pertinent literature, outlining the important models and theories underlying technology acceptance paradigms in education. Findings have demonstrated that informants in this study have identified the facilitators and inhibitors of e-learning previously recognised in prior research. ey also showed that students are ready to accept technology implementation and to shift to an e-learning model of education.
Article
One of the challenges in information retrieval is providing accurate answers to a user's question often expressed as uncertainty words. Most answers are based on a Syntactic approach rather than a Semantic analysis of the query. In this paper, our objective is to present a hybrid approach for a Semantic question answering retrieval system using Ontology Similarity and Fuzzy logic. We use a Fuzzy Co-clustering algorithm to retrieve the collection of documents based on Ontology Similarity. The Fuzzy Scale uses Fuzzy type-1 for documents and Fuzzy type-2 for words to prioritize answers. The objective of this work is to provide retrieval system with more accurate answers than non-fuzzy Semantic Ontology approach.
Book
During the last two decades, the idea of Semantic Web has received a great deal of attention. An extensive body of knowledge has emerged to describe technologies that seek to help us create and use aspects of the Semantic Web. Ontology and agent-based technologies are understood to be the two important technologies here. A large number of articles and a number of books exist to describe the use individually of the two technologies and the design of systems that use each of these technologies individually, but little focus has been given on how one can - sign systems that carryout integrated use of the two different technologies. In this book we describe ontology and agent-based systems individually, and highlight advantages of integration of the two different and complementary te- nologies. We also present a methodology that will guide us in the design of the - tegrated ontology-based multi-agent systems and illustrate this methodology on two use cases from the health and software engineering domain. This book is organized as follows: Chapter I, Current issues and the need for ontologies and agents, describes existing problems associated with uncontrollable information overload and explains how ontologies and agent-based systems can help address these - sues. Chapter II, Introduction to multi-agent systems, defines agents and their main characteristics and features including mobility, communications and collaboration between different agents. It also presents different types of agents on the basis of classifications done by different authors.