Enhancing just-in-time e-learning through machine learning on desktop context sensors

Abstract

The objective of novel e-learning strategies is to educate the learner during his actual work process. We focus on this new approach of in-place and in-time e-learning, which offers learning resources right in time the user is in need for it. A crucial factor for those modern task-oriented e-learning software is the user's context. To deliver learning resources to the user, which are both suitable and helpful with regards to the user's current work situation and his competencies, the application always has to consider the learner's actual work task, his environment, and history. In this paper, we present an architecture for the work task prediction, evaluate different machine learning algorithms in depth by their accuracy for that purpose and discuss the integration in our elearning environment. This validates the possible usage in real-world business scenarios. © Springer-Verlag Berlin Heidelberg 2007.