Techniques, Technologies and Patents Related to Intelligent Educational Systems - E-Infrastructures and Technologies for Lifelong Learning - page 12

Solutions from Artificial
Intelligence Research

comes an issue (Nesic, Gasevic & Jazayeri,
2008).
• Communication/collaboration with other
IESs. In a Web-based context, commu-
nication/collaboration among IESs may
offer advantages to learners (e.g. finding
additional learning resources (Avouris &
Solomos, 2001), data exchange for user
model initialization/updating and applica-
tion of appropriate instructional strategies
(Rishi & Govil, 2008)).

Various AI technologies/techniques are used in
IESs to achieve tasks concerning all stages of
an IES's life cycle (construction, operation, and
maintenance), all types of users (authors, tutors,
learners) and all its modules (domain knowledge,
user modeling, pedagogical module). Knowledge
representation and reasoning (KR&R) is of great
importance, since what is mainly needed is repre-
sentation of human reasoning (Russell & Norvig,
2009). Therefore, we briefly discuss issues related
to the following methods/techniques: structured
and relational schemes, rule-based reasoning,
case-based reasoning, neural networks, Bayesian
networks, fuzzy logic, constraint-based modeling,
genetic algorithms, reinforcement learning, hybrid
KR&R techniques, data mining and intelligent
agents.
Before proceeding, we give brief definitions
for the terms 'classification', 'clustering' and
'generalization', which refer to corresponding
types of tasks to be carried out. The term 'clas-

Table 1 outlines issues regarding IESs that
should be tackled.

SOLUTIONS AND
RECOMMENDATIONS

In this section, we provide solutions/recommen-
dations to several of the aforementioned issues
regarding IESs.

Table 1. Summary of issues regarding IESs that should be tackled

Choosing a representation scheme for domain knowledge
Domain knowledge structure creation
Learning unit creation
Provision of learning unit metadata
Maintenance of domain knowledge items
Support for domain knowledge construction/maintenance by multiple collaborating authors

Domain knowledge issues

Choosing which learner characteristics to record
Choosing the representation scheme for learner characteristics
Implementation of mechanisms to assess learner characteristics
Exploitation of user model data to provide useful information to tutors and learning content authors

User modeling issues

Deciding which tasks will be implemented
Pedagogical knowledge acquisition
Maintenance of pedagogical knowledge
Choosing intelligent techniques for implementing pedagogical tasks

Pedagogical module issues

Provision of authoring tools/facilities for IESs
The domain type to which the IES can be applied
Collaborative learning support
Support of natural language processing or dynamic natural language responses
Combination of IES and LMS functionality
Tutoring with simulations
Distributed learning content
Communication/collaboration with other IESs

General issues for IESs