|
Title:
|
ACHIEVING ADAPTIVE E-LEARNING USING CONDITION-ACTION RULES AND BAYESIANNETWORKS |
|
Author(s):
|
Sanghyun S. Jeon , Stanley Y. W. Su , Gilliean Lee |
|
ISBN:
|
978-972-8924-42-3 |
|
Editors:
|
Miguel Baptista Nunes and Maggie McPherson (series editors: Piet Kommers, Pedro Isaías and Nian-Shing Chen) |
|
Year:
|
2007 |
|
Edition:
|
V I, 2 |
|
Keywords:
|
Adaptive e-learning management system, Bayesian Network, Learning object, Event-triggered rule processing. |
|
Type:
|
Full Paper |
|
First Page:
|
359 |
|
Last Page:
|
366 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
Although Web-based learning management systems (LMSs) can deliver instructions to learners anywhere and anytime,
their one size fits all services do not meet the diverse needs of learners who have different backgrounds, goals, and
preferences. Recent research efforts on adaptive LMSs focus on the use of learners profiles and run-time progress data to
tailor instructions to meet the diverse needs of the learners. However, the profile information provided by learners and the
progress data collected or generated by a LMS are often incomplete, inaccurate, and/or contradictory. An adaptive LMS
must be able to accommodate these data uncertainties. This paper presents an implemented adaptive LMS called Gator ELearning
System (GELS), which uses an event-trigger-rule-based approach to process learning objects (LOs). Events are
posted at various stages of processing a learning object to trigger relevant rules to control the sequence, the control path,
and the presentation of instructional materials. GELS also use Bayesian Networks to model and evaluate probabilistically
the condition part of a rule that makes reference to profile and progress information. The integration of an event-triggerrule
(ETR) Server and a Bayesian Network component enables GELS to provide individualized instructions in the
presence of data uncertainties. |
|
|
|
|
|
|
