LEARNING TECHNOLOGY SYSTEM ARCHITECTURE BASED
ON AGENTS AND SEMANTIC WEB
Alejandro Canales and Rubén Peredo
Computer Science Research Center of National Polytechnic Institute, Col. Nueva Industrial Vallejo
Del. Gustavo A, Madero, D.F. 07738, Mexico City, Mexico
Keywords: Web-Based Education, Agent-Oriented Programming, IRLCOO.
Abstract: The paper presents a new Learning Technology System Architecture that is implemented using agents and
semantic Web. The architecture is divided into client and server parts to facilitate adaptivity in various
configurations such as online, offline and mobile scenarios. The implementation of this approach to
architecture is discusses in SiDeC (authoring tool) and Evaluation System, which are researching,
evaluating and developing an integrated system for Web-Based Education, with powerful adaptivity for the
management, authoring, delivery and monitoring of such material.
1 INTRODUCTION
The use of Web-Based Education (WBE) as a mode
of study is due to the increase in the number of
students and limited learning content resources
available to meet a wide range of personal needs,
backgrounds, expectations, skills, levels, etc. In this
way, the purpose of the delivery process is very
important, because it means to produce learning
content and to present it to the learner in multimedia
form. Nowadays, there are approaches over this
process that focus on new paradigms to produce and
deliver quality content for online learning
experiences, such as a special type of labeled
materials called Intelligent Reusable Learning
Components Object Oriented (IRLCOO), developed
by Peredo et al (2005).
IRLCOO are part of a new Learning Technology
System Architecture (LTSA) based on IEEE 1484
specification (IEEE, 2001) and open standards such
as XML (XML, 2006) as a bar coding system and to
make sure that the learning content is interoperable,
the Global IMS Learning Consortium (IMS, 2005),
Advanced Distributed Learning (ADL), and
SCORM (ADL, 2006). This paper is organized as
follows: in Section 2, LTSA and IRLCOO are
described; in Section 3 y 4, the authoring system
called SiDeC and the Evaluation System are
presented; finally, the conclusions are discussed.
2 LEARNING TECHNOLOGY
SYSTEM ARCHITECTURE
Our LTSA is based on layer 3 of IEEE 1484
specification. This architecture is presented in figure
1, and consists in four processes manage by agents:
Learner Entity, Evaluation, Coach, and Delivery
process; two stores: Learner Records and Learning
Resources; and fourteen information workflows.
First, the Coach process has been divided in two
subprocesses: Coach and Virtual Coach. The reason
is because we considered that this process has to
adapt to the learners’ individual needs in a quick
way during the learning process. For this, some
decisions over sequence, activities, examples, etc.,
can be made manually for the coach but in others
cases these decisions can be made automatically for
the virtual coach.
Figure 1: Learning Agents System.
Briefly, the overall operation has the following
form: (1) the learning styles, strategies, methods,
etc., are negotiated among the learner and other
stakeholders and are communicated as learning
127
Canales A. and Peredo R. (2008).
LEARNING TECHNOLOGY SYSTEM ARCHITECTURE BASED ON AGENTS AND SEMANTIC WEB.
In Proceedings of the Tenth International Conference on Enterprise Information Systems - SAIC, pages 127-132
DOI: 10.5220/0001698301270132
Copyright
c
SciTePress
preferences; (2, new proposal) the learner
information (behavior inside the course, e.g.,
trajectory, times, nomadicity, etc.) is stored in the
Learner Records; (3) the learner is observed and
evaluated in the context of multimedia interactions;
(4) the Evaluation produces assessments and/or
learner information; (5) the learner information
(keyboard clicks, mouse clicks, voice response,
choices, written responses, etc., all over learner’s
evaluation) is stored in the learner history data-base;
(6) the Coach reviews the learner's assessment and
learner information, such as preferences, past
performance history, and, possibly, future learning
objectives; (7, new proposal) the Virtual Coach
reviews the learner’s behavior and learner
information, and automatic and smartly he makes
dynamic modifications on the course sequence
(personalized to learner’s needs) based on the
learning process design; (8) the Coach/Virtual
Coach searches the learning resources, via query and
catalog info, for appropriate learning content; (9) the
Coach/Virtual Coach extracts the locators (e.g.,
URLs) from the available catalog info and passes the
locators to the delivery process, e.g., a lesson plan or
pointers to content; and (10) the Delivery process
extracts the learning content and the learner
information from the Learning Resources and the
Learner Records respectively, based on locators, and
transforms the learning content to an interactive and
adaptive multimedia presentation to the learner.
2.1 IRLCOO Platform
IRLCOO were developed with Flash. Flash is an
integrator of media and have a powerful
programming language denominated ActionScript
3.0 (Adobe, 2007). This language is completely
Object Oriented and enables the design of learning
components that allows multimedia content of side
client. At Run-Time, the components load media
objects and offer a programmable and adaptive
environment to the student's necessities. Flash
already has Smart Clips for the learning elements
denominated Learning Interactions. The aim is to
generate a multimedia library of IRLCOO for WBE
systems with the purpose to separate the content
from the control. Thus, the components use different
levels of code inside the Flash Player. With this
structure, it is possible to generate specialized
components which are small, reusable, and suitable
to integrate them inside a bigger component at Run-
Time by Delivery process. The liberation of
ActionScript version 3.0 inside Adobe Flash©
allows the implementation of the Object Oriented
paradigm. With these facilities IRLCOO are tailored
to the learners’ needs. In addition, this IRLCOO
development platform owns certain communication
functionalities inside the Application Programming
Interface with LMS, Multi-Agent System (MAS),
and different frameworks, as AJAX (Crane, 2006),
Hibernate (Peak, 2006), Struts (Holmes, 2004), etc.,
and dynamic load of assets in Run-Time.
IRLCOO are meta-labeled with the purpose of
complete a similar function as the product bar codes,
which are used to identify the products and to
determine certain characteristics specify of
themselves. This contrast is made with the meta-
labeled Resource Description Framework (RDF-
XML) (RDF, 2004), which allows enabling certain
grade inferences on the materials by means of the
Semantic Web Platform.
2.2 Communication between IRLCOO
and Web Services
The Web Service (WS) standards enable a set of
basic interactions required in a Service Oriented
Architecture (SOA). WS allow access to
functionality via the Web using a set of open
standards that make the interaction independent of
implementation aspects such as the operating system
platform and the programming language used.
ActionScript 3.0 adds the component
WebServiceConnector to connect to WS from the
IRLCOO. The WebServiceConnector component
enables the access to remote methods offered by a
LMS through SOAP protocol. This gives to a WS
the ability to accept parameters and return a result to
the script, in other words, it is possible to access and
join data between public or own WS and the
IRLCOO. It is possible to reduce the programming
time, since a simple instance of the
WebServiceConnector component is used to make
multiple calls to the same functionality within the
LMS. The components discover and invoke WS
using SOAP and UDDI, via middleware and a
JUDDI server (JUDDI, 2005). Placing a Run-Time
layer between a WS client and server dramatically
increases the options for writing smarter, more
dynamic clients. Reducing the needs for hard-coded
dependencies within WS clients (see figure 2). It is
only necessary to use different instances for each
one of the different functionalities. WS can be
unloaded using the component and deployed within
an IRLCOO.
ICEIS 2008 - International Conference on Enterprise Information Systems
128
Figure 2: Web service architecture with IRLCOO.
2.3 Mapping Object/Relational
Hibernate is a powerful, high performance
object/relational persistence and query service.
Hibernate lets develop persistent classes following
object-oriented idiom - including association,
inheritance, polymorphism, composition, and
collections (Hibernate, 2005).
According with figure 1, the “Learner Info
(new)” flow is implemented with Hibernate.
Creating the InfrastructureException class for
Hibernate Errors and Exceptions, this class is used to
manipulate exceptions in the Hibernate code. Later
creating the HibernateUtil class is possible to initiate
Hibernate and get access to a session of the
persistence and transaction operations.
The StudentDAO class encapsulates all
functionalities necessary to persist student’s registers
data in the database using Hibernate. Thus, the
action class will be responsible by get information
from student, pass to persist for the DAO class and
return success or error to the user. Mixing Struts and
Hibernate and creating the StudentAction class, this
class is an adapter between the contents of an
incoming HTTP request and the corresponding
business logic that should be executed by this
process request. The controller (RequestProcessor)
will select an appropriate Action for each request,
create an instance, and call the execute method. The
database configuration in Hibernate is made trough
of the hibernate.cfg.xml file. The mapping between
Student Class and database is carried out in the
Student.hbm.xml file.
3 SiDeC
In order to facilitate the development of learning
content, it was built an authoring system called
SiDeC (Sistema de Desarrollo de eCursos - eCourses
Development System). SiDeC is a system based on
LTSA to facilitate the authoring content to the tutors
who are not willing for handling multimedia
applications. In addition, the Structure and Package
of content multimedia is achieved by the use of
IRLCOO, as the lowest level of content granularity.
SiDeC is used to construct Web-based
courseware from the stored IRLCOO (Learning
Resources), besides enhancing the courseware with
various authoring tools. Developers choose one of
the SiDeC lesson templates and specify the desired
components to be used in each item. At this moment,
the SiDeC lesson templates are based on the
cognitive theory of Conceptual Maps (CM), but in
the future we will consider others theories such as:
Based-Problems Learning (BPL), the cases method,
etc.
SiDeC has a metadata tool that supports the
generation of IRLCOO to provide online courses.
This courseware estimates learners’ metrics with the
purpose to tailor their learning experiences.
Furthermore, the IRLCOO offer a friendly interface
and flexible functionality. These deliverables are
compliance with the specifications of the IRLCOO
and with learning items of SCORM 1.2 Models
(Content Aggregation, Sequencing and Navigation,
and Run Time Environment) (ADL, 2006). Metadata
represent the specific description of the component
and its contents, such as: title, description,
keywords, learning objectives, item type, and rights
of use. The metadata tool provides templates for
entering metadata and storing each component in the
SiDeC or another IMS/IEEE standard repository.
In figure 3, the SiDeC implements the CM as a
navigation map or instructional and learning strategy
allowing to the learner to interact with content
objects along the learning experiences. These
experiences follow an instructional-teaching
strategy. These kinds of strategies carry out
modifications of the learning content structure. Such
modifications are done by the designer of the
learning experience with the objective of provide
significant learning and to teach the learners how to
think (Díaz-Barriga, 2002). The learning content can
be interpreted in a Learning Content Tree.
Figure 3: Learning content generated for the SiDeC.
LEARNING TECHNOLOGY SYSTEM ARCHITECTURE BASED ON AGENTS AND SEMANTIC WEB
129
3.1 Communication between IRLCOO
and LMS
Our communication model uses an asynchronous
mode in Run-Time Environment (RTE) and joins to
LMS communication API of ADL (ADL, 2006),
AJAX (Asynchronous JavaScript And XML)
(Crane, 2006) and Struts Framework (Holmes, 2004)
for its implementation. The LMS communication
API of ADL consists of a collection of standard
methods to let the Client to communicate with the
LMS. The browser-based communication model is
depicted in figure 4.
Figure 4: Communication model between IRLCOO, LMS
and AJAX and Struts Framework.
According to figure 4, the communication model
starts: (1) when an IRLCOO generates an event. (2)
Form the browser interface is made a JavaScript call
to the function FileName_DoFSCommand
(command,args), which handles all the FSCommand
messages from IRLCOO, LMS communication API,
and AJAX and Struts methods.
The communication with the LMS starts when:
(i) the standard methods call to the Communication
Adapter (written in JavaScript). (ii) The
communication adapter implements the bidirectional
communication ADL´s API between the Client and
the LMS. (iii) The LMS realizes the query-response
handling and the business logic. The purpose of
establishing a common data model is to make sure
that a defined set of information about content can
be tracked by different LMS environments. If, for
example, it is determined that tracking a student’s
score is a general requirement, then it is necessary to
establish a common way for content to report scores
and for LMS environments to process such
information (for interoperability and reuse of such
content). If every chunk of content used its own
unique scoring representation learning management
systems would not know how to receive, store or
process such information. These definitions are
derived from the IEEE Standard for Computer-
Managed Instruction document (P1484.11), which
originated within the Aviation Industry CBT
Committee (AICC) (IEEE, 2005).
The communication with AJAX and Struts
Framework begins when AJAX-Struts method is
called. (3) An instance of the XMLHttpRequest
object is created. Using the open() method, the call
is set up, the URL is set along with the desired
HTTP method, typically GET or POST. The request
is actually triggered via a call to the send() method.
(4) A request is made to the server, this might be
a call to a servlet or any server side technique. (5)
The Controller is a servlet which coordinates all
applications activities, such as: reception of user
data, data validations, and control flow. The
Controller is configured for a XML file. The
Controller calls to perform method of Action, it
passes to this method the data values and the Action
reviews the characteristic data that correspond to the
Model. The business objects (JavaBeans) realize the
business logic, (5) usually a database access by
Hibernate. The Action sends the response to the
Controller. The Controller reroutes and generates the
interface for the results to the View (JSPs). The
View makes the query to the Business objects based
on the correspondent interface. (6) The request is
returned to the browser. The ContentType is set to
text/xml, the XMLHttpRequest object can process
results only of the text/ html type. In more complex
instances, the response might be quite involved and
include JavaScript, DOM manipulation, or other
technologies. The XMLHttpRequest object calls the
function callback() when the processing returns.
This function checks the readyState property on the
XMLHttpRequest object and then looks at the status
code returned from the server. (7) Provided
everything is as expected, the callback() function
sends HTML code and it does something interesting
on the client, i.e. advanced dynamic sequence.
This communication model provides new wide
perspectives for the WBE systems development,
because it provides the capabilities of communica-
tion, interaction, interoperability, security, and
reusability, between different technologies. For
example, the LMS communication API allows us to
make standard database queries of learners’
information such as personal information, scores,
assigned courses, trajectory, etc. While the
communication with AJAX and Struts Framework
ICEIS 2008 - International Conference on Enterprise Information Systems
130
provides the capability of modify the learner’s
trajectory according to variables from the learner
records in RTE (advanced dynamic sequence),
components management (IRLCOO) – remember
that these components are built and programming
with XML – then, this model provides the way to
write, load, change and erase XML files in the
Server side.
4 EVALUATION SYSTEM
The Evaluation System for WBE is designed under
the same philosophy used for the SiDeC. The
functionality of the Evaluation System lays on the
analysis of the learner’s profile, which is built during
the teaching-learning experiences. The profile is
based on metrics that elicited from the learner’s
behavior at Run-Time. These measures are stored
into the learner records that compose the profile. The
generation of new sequences of courses is in
function of the results obtained, besides the account
of the adaptation level.
The Evaluation System combines IRLCOOs,
additional meta-labels, and a Java Agent platform.
Also, some technologies of the Artificial Intelligence
field are considered in order to recreate a Semantic
Web environment. Semantic Web aims for assisting
human users to achieve their online activities.
In resume, the components and operation of the
SiDeC and Evaluation System are outlined in figure
5. Basically the Evaluation System is fulfilled
through two phases. The first phase is supported by
the LMS, and is devoted to present the course and its
structure. All the actions are registered and the
presentation of the contents is realized with
IRLCOO content. The evaluations are done by
evaluating IRLCOO and in some cases by simulators
based on IRLCOO. These processes are deployed by
the Framework of Servlets, JSPs and JavaBeans.
The second phase analyzes the learner's records
carried out by the Server based on JADE MAS. This
agent platform owns seven agents: Snooper, Buffer,
Learner, Evaluation, Delivering, Coach, and Info.
The fundamental idea is to automate the learner's
analysis through the Coach/Virtual Coach, and to
give partial results that can be useful for the learner's
final instruction. These agents are implemented as
Java-Beans programs, which are embedded in the
applications running both at the client and server
sides. The Snooper Agent works as a trigger by
means of the INFORM performative, which
activates the MAS server’s part. This agent is
deployed into a Java Server Page that uses a
JavaBean. During the lesson or once evaluation is
finished, the graphical user interface activates the
Snooper Agent and sends it the behavior or
evaluation metrics (using Agents Communications
Language or FIPA ACL (FIPA, 2001) to be
analyzed at the server-side of the MAS. The Snooper
Agent activates the system, whereas the Buffer
Agent manages the connection and all the messages
from the client. Both tasks are buffered and send
them to the Coach Agent. Then the Coach Agent
requests to the learner records for the preferences
learner, trajectory, previous learner monitoring
information, etc. The Coach Agents analyzes this
information to determine if the learner needs help. If
this situation is true, the Coach Agent requests to the
learning resources the needful learning content
(URLs) and it sends the learning contents (URLs) to
the Delivery Agent. The Delivery Agent sends the
learning content to the Learner and Evaluation
Agents for its presentation. These agents employ the
dynamic sequencing to change the course or
assessment sequence. The sequencing is defined for
the instructional strategy based on CM and it
employs the SCORM Sequencing/Navigation. Once
the necessary information is received (sequence,
kind of IRLCOO and localization, etc.), this is
represented as a string, which is constructed
dynamically by the rule-based inference engine
known as JENA (JENA, 2006) and JOSEKI server
(JOSEKI, 2006), to generate dynamic feedback.
Figure 5: Semantic Web Platform for WBE.
4.1 Semantic Web Platform
The overall architecture of Semantic Web Platform,
which includes three basic aspects (see figure 5):
LEARNING TECHNOLOGY SYSTEM ARCHITECTURE BASED ON AGENTS AND SEMANTIC WEB
131
1. The query engine receives queries and
answers them by checking the content of the
databases that were filled by info agent and
inference engine.
2. The database manager is the backbone of
the entire systems. It receives facts from the info
agent, exchanges facts as input and output with the
inferen-ce engine, and provides facts to the query
engine.
3. The inference engine use facts and
ontologies to derive additional factual knowledge
that is only provided implicated. It frees knowledge
providers from the burden of specifying each fact
explicitly.
Again, ontologies are the overall structuring
principle. The info agent uses them to extracts facts,
the inference engine to infer facts, the database
manager to structure the database, and query engine
to provide help in formulating queries.
JENA was selected as the inference engine. It is
a Java framework for building Semantic Web
applications. It provides a programmatic environ-
ment for RDF, RDFS, OWL, SPARQL and includes
a rule-based inference engine (JENA, 2006).
While JOSEKI was selected as Web API and
server. It is an HTTP and SOAP engine supports the
SPARQL Protocol and the SPARQL RDF Query
language. SPARQL is developed by the W3C RDF
Data Access Working Group (JOSEKI, 2006).
5 CONCLUSIONS
LTSA, IRLCOO and Semantic Web Platform allow
developing authoring and evaluation systems to
create adaptive and intelligent WBE. Our approach
focus on: reusability, accessibility, durability, and,
interoperability of the learning contents.
The communication model composes for the
LMS communication API, AJAX and Struts
Framework, IRLCOO, WS, Semantic Web, and
JUDDI. It provides new development capabilities for
WBE systems, because their integrant technologies
are complementary. SiDeC and the Evaluation
System were developed under this model to help in
the automation and reduce of the complexity of the
learning content process.
The incorporation of Web Semantic Platforms
helps us to create intelligent and adaptive systems
(bidirectional communication), according to the
users’ needs.
ACKNOWLEDGEMENTS
Authors of this paper would like to thank the
Instituto Politécnico Nacional (IPN), Centro de
Investigación en Computación (CIC) for the partial
support for this project within the project IPN
20071166.
REFERENCES
ADL (2006). Advanced Distributed Learning Consortium.
Retrieved January 24, 2006 from
http://www.adlnet.org
Adobe (2007). Adobe© Flash©. Retrieved February 26,
2007 from http://www.adobe.com
Crane. D., Pascarello, E., James, D., 2006. Ajax in Action,
Manning Publications, 1
st
edition.
Díaz-Barriga, F., 2002. Educational strategies for a
significant learning, Mc Graw Hill Publication,
Second edition.
FIPA (2001). FIPA ACL message structure specification,
XC00061. Retrieved August 20, 2005 from
http://www.fipa.org/
Hibernate (2005). Hibernate. Retrieved October 11, 2006
from http://www.hibernate.org/
Holmes, J., 2004. Struts: The Complete Reference, Mc
Graw Hill – Osborne Publications, 1
st
edition.
IEEE (2001). IEEE 1484.1/D9 LTSA: Draft Standard for
Learning Technology - Learning Technology Systems
Architecture. Retrieved December 8, 2003 from
http://ieee.ltsc.org/wg1
IEEE (2005). IEEE Learning Technology Standards
Committee. Retrieved December 10, 2005 from
http://ltsc.ieee.org/wg11/index.html
IMS (2005). Global IMS Consortium. Retrieved March
22, 2005 from http://www.imsproject.org
JENA (2006). Jena – A Semantic Web Framework for
Java. Retrieved June 9, 2006 from:
http://jena.sourceforge.net/
JOSEKI (2006). Joseki - A SPARQL Server for Jena.
Retrieved June 9, 2006 from: http://www.joseki.org/
JUDDI (2005). Apache Web Service Project. Retrieved
November 4, 2006 from http://ws.apache.org/juddi/
Passin, T., 2004. Explorer’s Guide to Semantic Web.
Manning Publications Co., 1
st
edition.
Peak, P., Heudecker, N., 2006. Hibernate Quickly,
Manning Publications, 1
st
edition.
Peredo, R., Balladares, L., Sheremetov, L., 2005.
Development of intelligent reusable learning objects
for web-based education systems. Expert Systems with
Applications. 28(2). 273-283.
RDF (2004), Resource Description Framework. Retrieved
January 15, 2005 from http://www.w3.org/RDF/
XML (2006). eXtensible Markup Language specifications.
Retrieved January 7, 2007 from
http://www.w3.org/XML.
ICEIS 2008 - International Conference on Enterprise Information Systems
132