Learner typing for electronic learning systems

e-Learning systems will provide new pathways to knowledge and skills acquisition. To increase their quality using a parametric learner model is proposed. Tailoring the learning material presented to the respective parameter values may fit this material to the actual learner's needs. We use a stereotype user model. Our learner characteristics complement the IMS UP. We first represent learners by points in a learner space. Then we propose increasingly generic ways of grouping together such points to form learner types. The less generic one models learner types using convex regions in the learner space. The more generic one models learner types using characteristic functions. While the latter approach allows for more flexible learner type combination the former approach allows for convincing definition of initial learner types. Combining both approaches thus may result in a reasonable number of useful learner types.

1. Introduction

e-learning systems can be considered as specific Web-based information systems with a focus on the provision of knowledge to learners if access to these systems is given via Web. It appears that in a high percentage of the currently available e-learning systems, the learner has to follow a predefined sequence of learning steps. Such sequence often reflects the content designer's preferred approach to mastering the offered curriculum. To some extent it follows the principles of a didactic preparation like summarized in [6]. However, helping learners better should allow these not to follow a predefined path and in doing so accounting for, e.g. preferred learning style, personal characteristics, learning task related characteristics and preferred examination styles. e-learning systems should restrain learners as little as possible. Rather as much and diverse learning-and examination support as possible should be offered.