=Paper=
{{Paper
|id=None
|storemode=property
|title=A Java Framework for Developing Intelligent Tutoring Systems
|pdfUrl=https://ceur-ws.org/Vol-692/paper2.pdf
|volume=Vol-692
}}
==A Java Framework for Developing Intelligent Tutoring Systems==
https://ceur-ws.org/Vol-692/paper2.pdf
A Java Framework for
Developing Intelligent
Tutoring Systems
Julian Arthur and Colin Depradine
Department of Computer Science, Mathematics
and Physics
The University of the West Indies
Bridgetown, Barbados
julian.arthur@hotmail.com
colin.depradine@cavehill.uwi.edu
ABSTRACT
An intelligent tutoring system (ITS) is software that provides an interactive learning envi-
ronment and can adapt its instruction depending on user needs. The development of these
systems is expensive due to significant resource requirements. To ease the development pro-
cess of these systems, technologies such as authoring tools and shells are sometimes used.
In this paper a new java framework is proposed to be implemented as the foundation of a
new authoring tool for intelligent tutoring systems (ITSs).
1 Introduction
The development of ITSs is expensive due to the large amount of expertise needed
and the long development times [1]. The development team for a single ITS can con-
sist of numerous experienced programmers as well as experts in number of varying
research areas, such as educational theory and domain knowledge design.
The potential benefits when using ITSs are great but unfortunately the difficulty in
producing these systems has limited both access and acceptance of them. Therefore,
efficient and effective authoring technologies are required to assist in creating ITSs.
This paper presents the motivation for creating this new java framework and intro-
duces the fundamental architecture of the framework.
2 Problem
The conceptual designs of ITSs are modular but unfortunately, typically the final im-
plementations of these systems are quite monolithic in structure [2] which can be
attributed to the functionality required for effective instruction of specific material.
The difficulty in efficiently and effectively creating ITSs has also reduced their acces-
sibility and acceptance in the traditional classroom.
1
PPPJ’10 WiP Poster Abstract
3 Motivation
To provide a java-based framework that will encourage both modular design of the
ITS and reuse of ITS components. Another reason for this work is to provide a sen-
sible foundation for an authoring tool that can be used to simplify and improve ITS
development productivity.
4 Goals and Objectives
The first goal is to reduce the development cost and implementation times of ITS soft-
ware by providing a comprehensive ITS authoring tool (includes framework). The
secondary goal is to establish a public test bed for research into ITS design and im-
plementation by providing the framework as an open source tool.
The objectives are that participating developers using this framework and authoring
tool will:
• Act as a catalyst to improve both the framework and tool.
• Create an active developer community in this research area thereby benefitting
from peer interaction.
• Create an industry standard for the design and implementation of ITSs.
5 ITS System Design
The java language was chosen as the primary language for this project due to its
object-oriented paradigm and due to the java community’s open source nature.
The architecture for this framework consists of seven primary modules as shown in
figure 1. The primary modules in the ITS architecture are (1) the distributed learning
module, (2) the pedagogical module, (3) the domain knowledge module, (4) the stu-
dent model, (5) the expert model , (6) the communications module and (7) the control
module.
Figure 1: The fundamental architecture representing the framework
2
PPPJ’10 WiP Poster Abstract
The conceptual flow of information between modules within an ITS can be seen
below in figure 2.
Figure 2: Conceptual flow of communication between components
5.1 Student Model
The student model holds unique user information which represents the user’s un-
derstanding of the material taught by the ITS. A user’s actions are monitored by the
system and the model is updated appropriately to track a student’s performance.
5.2 Expert Model
The expert model contains the information that is taught to the user. However, it is
more than just a representation of the data; it is a model of how someone skilled in a
particular domain uses the knowledge. An expert model can be used to pinpoint user
difficulties by comparing the learner’s solution to the expert’s solution.
5.3 Domain Knowledge
This component contains all the information that the tutoring system will teach to
an ITS user. The representation of the domain knowledge is critical since the other
components which must access this module should be able to do so efficiently.
5.4 Pedagogical Module
The pedagogical module performs a number of tasks such as determining when to
review a topic and deciding when to present a new topic. This module uses input
provided by the student model to adapt to a user’s specific needs.
5.5 Communications Module
The communications module is the user interface of the ITS. All interactions between
the user and the ITS are handled by this component. This module handles the pre-
sentation of the taught material ensuring it is presented in an appropriate way.
5.6 Distributed Learning Module
The distributed learning module is a nonconventional component which has been
added and it is used to provide various networking functionality in the ITS environ-
ment.
3
PPPJ’10 WiP Poster Abstract
5.7 Control Module
This module acts as the central control for the primary modules of the ITS. All com-
munication between modules is relayed via this component.
6 Conclusion
We have provided a brief explanation of what research area we are exploring, the
limitations that currently exist in ITS implementations, our goals for the future of
this research and present the architecture currently adopted in the framework and
authoring tool.
References
[1] T. Murray. Authoring intelligent tutoring systems: An analysis of the state of the
art. International Journal of Artificial Intelligence in Education, 10:98 – 129, 1999.
[2] T. Murray. An overview of intelligent tutoring system authoring tools: Updated
analysis of the state of the art. In T. Murray, S. Blessing, and S. Ainsworth, editors,
Authoring Tools for Advanced Technology Learning Environments, pages 493 – 546.
Kluwer Academic Publishers, Dordrecht, 2003.
[3] V. Devedzic and A. Harrer. Common patterns in its architectures. Kunstliche
Intelligenz, 18:17 – 21, 2004.
4