Title: Adaptive e-learning using ECpAA rules, Bayesian models, and group profile and performance data

Authors: Sanghyun S. Jeon, Stanley Y.W. Su

Addresses: University of Florida, Database Systems R&D Center, Gainesville, FL 32611, USA. ' University of Florida, Database Systems R&D Center, Gainesville, FL 32611, USA

Abstract: An e-learning system must be capable of gathering and correctly evaluating a learner|s profile and performance data in order to deliver individualised instruction to the learner. However, the learner|s data can be incomplete, inaccurate and/or contradictory. They can also be correlated. This paper aims to alleviate these data problems by evaluating the data of each new learner probabilistically based on the data of earlier learners. Our probabilistic rule model allows our system to apply adaptation rules to examine learners| data at various stages of a learning activity, and determine the suitable actions to take to personalise the instruction. Adaptation rules are processed by a rule engine and a Bayesian model processor to achieve adaptive content search and selection, adaptive processing of learning objects, and continuous improvement on the accuracy of learner evaluation.

Keywords: adaptive e-learning; electronic learning; rule models; Bayesian networks; group profiles; personalised learning; composite learning objects; ECpAA; event-condition; probability-action; alternative action; performance data; data evaluation; learner profiles; incomplete data; inaccurate data; contradictory data; probabilistic rules; adaptation rules; learning activities; rule engines; adaptive content search; adaptive content selection; adaptive processing; continuous improvement; higher education; universities; accuracy; learner evaluation; USA; United States; University of Florida; Food and Drug Administration; learning technology.

DOI: 10.1504/IJLT.2010.038776

International Journal of Learning Technology, 2010 Vol.5 No.4, pp.415 - 434

Published online: 02 Mar 2011 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article